TR 10002801v010401p

ETSI TR 100 028-1 V1.4.
1 (2001-12)
Technical Report
Electromagnetic compatibility
and Radio spectrum Matters (ERM);
Uncertainties in the measurement
of mobile radio equipment characteristics;
Part 1

2 ETSI TR 100 028-1 V1.4.1 (2001-12)
Reference
RTR/ERM-RP02-058-1
Keywords
measurement uncertainty, mobile, radio
ETSI
650 Route des Lucioles

F-06921 Sophia Antipolis Cedex - FRANCE
Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16
Siret N° 348 623 562 00017 - NAF 742 C

Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88
Important notice
Individual copies of the present document can be downloaded from:

http://www.etsi.org
The present document may be made available in more than one electronic version or in print. In any case of existing or
perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF).
In case of dispute, the reference shall be the printing on ETSI printers of the PDF version kept on a specific network drive
within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, send your comment to:
[email protected]
Copyright Notification
No part may be reproduced except as authorized by written permission.

The copyright and the foregoing restriction extend to reproduction in all media.
© European Telecommunications Standards Institute 2001.

All rights reserved.
ETSI
3 ETSI TR 100 028-1 V1.4.1 (2001-12)
Contents
Intellectual Property Rights ................................................................................................................................8
Foreword.............................................................................................................................................................8
Introduction ........................................................................................................................................................8
1 Scope ......................................................................................................................................................10
2 References ..............................................................................................................................................10
3 Definitions, symbols and abbreviations .................................................................................................10
3.1 Definitions........................................................................................................................................................10
3.2 Symbols............................................................................................................................................................14
3.3 Abbreviations ...................................................................................................................................................17
4 Introduction to measurement uncertainty...............................................................................................17
4.1 Background to measurement uncertainty .........................................................................................................17
4.1.1 Commonly used terms ................................................................................................................................17
4.1.2 Assessment of upper and lower uncertainty bounds ...................................................................................18
4.1.3 Combination of rectangular distributions ...................................................................................................19
4.1.4 Main contributors to uncertainty.................................................................................................................21
4.1.5 Other contributors .......................................................................................................................................21
4.2 Evaluation of individual uncertainty components ............................................................................................22
4.2.1 Evaluation of type A uncertainties..............................................................................................................23
4.2.2 Evaluation of type B uncertainties ..............................................................................................................23
4.2.3 Uncertainties relating to influence quantities..............................................................................................24
4.3 Methods of evaluation of overall measurement uncertainty.............................................................................24
4.4 Summary ..........................................................................................................................................................25
4.5 Overview of the approach of the present document .........................................................................................25
5 Analysis of measurement uncertainty ....................................................................................................25
5.1 The BIPM method ............................................................................................................................................26
5.1.1 Type A uncertainties and their evaluation ..................................................................................................26
5.1.2 Type B uncertainties and their evaluation...................................................................................................26
5.2 Combining individual standard uncertainties in different units........................................................................27
5.3 Calculation of the expanded uncertainty values and Student's t-distribution ...................................................29
5.3.1 Student's t-distribution ................................................................................................................................29
5.3.2 Expanded uncertainties ...............................................................................................................................29
5.4 Combining standard uncertainties of different parameters, where their influence on each other is
dependant on the EUT (influence quantities) ...................................................................................................30
5.5 Uncertainties and randomness ..........................................................................................................................32
5.6 Summary of the recommended approach .........................................................................................................33
6 Examples of uncertainty calculations specific to radio equipment ........................................................34
6.1 Mismatch..........................................................................................................................................................34
6.2 Attenuation measurement.................................................................................................................................35
6.3 Calculation involving a dependency function ..................................................................................................37
6.4 Measurement of carrier power..........................................................................................................................39
6.4.1 Measurement set-up ....................................................................................................................................39
6.4.2 Method of measurement .............................................................................................................................39
6.4.3 Power meter and sensor module .................................................................................................................39
6.4.4 Attenuator and cabling network..................................................................................................................40
6.4.4.1 Reference measurement ........................................................................................................................41
6.4.4.2 The cable and the 10 dB power attenuator ............................................................................................42
6.4.4.3 The 20 dB attenuator.............................................................................................................................44
6.4.4.4 Instrumentation .....................................................................................................................................45
6.4.4.5 Power and temperature influences ........................................................................................................45
6.4.4.6 Collecting terms ....................................................................................................................................45
6.4.5 Mismatch during measurement...................................................................................................................46
6.4.6 Influence quantities.....................................................................................................................................47
ETSI
4 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.4.7 Random.......................................................................................................................................................47
6.4.8 Expanded uncertainty .................................................................................................................................48
6.5 Uncertainty calculation for measurement of a receiver (Third order intermodulation) ....................................49
6.5.1 Noise behaviour in different receiver configurations..................................................................................49
6.5.2 Sensitivity measurement .............................................................................................................................50
6.5.3 Interference immunity measurements.........................................................................................................51
6.5.4 Blocking and spurious response measurements..........................................................................................51
6.5.5 Third order intermodulation........................................................................................................................51
6.5.5.1 Measurement of third order intermodulation ........................................................................................51
6.5.5.2 Uncertainties involved in the measurement ..........................................................................................52
6.5.5.2.1 Signal level uncertainty of the two unwanted signals......................................................................53
6.5.5.2.2 Signal level uncertainty of the wanted signal ..................................................................................54
6.5.5.3 Analogue speech (SINAD) measurement uncertainty...........................................................................54
6.5.5.4 BER and message acceptance measurement uncertainty ......................................................................54
6.5.5.5 Other methods of measuring third order intermodulation .....................................................................54
6.6 Uncertainty in measuring continuous bit streams.............................................................................................55
6.6.1 General........................................................................................................................................................55
6.6.2 Statistics involved in the measurement.......................................................................................................55
6.6.3 Calculation of uncertainty limits when the distribution characterizing the combined standard
uncertainty cannot be assumed to be a Normal distribution .......................................................................56
6.6.4 BER dependency functions.........................................................................................................................58
6.6.4.1 Coherent data communications .............................................................................................................59
6.6.4.2 Coherent data communications (direct modulation) .............................................................................59
6.6.4.3 Coherent data communications (sub carrier modulation)......................................................................60
6.6.4.4 Non coherent data communication........................................................................................................61
6.6.4.5 Non coherent data communications (direct modulation) ......................................................................61
6.6.4.6 Non coherent data communications (sub carrier modulation)...............................................................63
6.6.5 Effect of BER on the RF level uncertainty .................................................................................................63
6.6.5.1 BER at a specified RF level ..................................................................................................................64
6.6.6 Limitations in the applicability of BER uncertainty calculations ...............................................................67
6.7 Uncertainty in measuring messages .................................................................................................................67
6.7.1 General........................................................................................................................................................67
6.7.2 Statistics involved in the measurement.......................................................................................................68
6.7.3 Analysis of the situation where the up down method results in a shift between two levels........................69
6.7.4 Detailed example of uncertainty in measuring messages............................................................................69
6.8 Examples of measurement uncertainty analysis (Free Field Test Sites)...........................................................72
6.8.1 Introduction.................................................................................................................................................72
6.8.2 Example 1: Verification procedure .............................................................................................................73
6.8.2.1 Uncertainty contributions: Stage 1: Direct attenuation measurement ...................................................73
6.8.2.1.1 Contributions from the mismatch components ................................................................................74
6.8.2.1.2 Contributions from individual components .....................................................................................76
6.8.2.1.3 Contribution from the random component ......................................................................................78
6.8.2.1.4 Summary table of contributory components....................................................................................79
6.8.2.2 Uncertainty contributions: Stage 2: Radiated attenuation measurement ...............................................79
6.8.2.2.2 Contributions from individual components .....................................................................................83
6.8.2.2.5 Expanded uncertainty for the verification procedure.......................................................................91
6.8.3 Example 2: Measurement of a transmitter parameter (spurious emission) .................................................91
6.8.3.1 Uncertainty contributions: Stage 1: EUT measurement ........................................................................91
6.8.3.1.2 Contributions from the individual components ...............................................................................92
6.8.3.2 Uncertainty contributions: Stage 2: Substitution measurement ............................................................99
6.8.3.2.2 Contributions from the individual components .............................................................................101
6.8.3.2.3 Contribution from the random component ....................................................................................107
6.8.3.2.4 Summary table of contributory components..................................................................................108
6.8.3.2.5 Expanded uncertainty for the spurious emission test.....................................................................108
6.8.4 Example 3: Measurement of a receiver parameter (Sensitivity) ...............................................................109
ETSI
5 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.4.1 Uncertainty contributions: Stage 1: Transform Factor measurement..................................................109

6.8.4.1.1 Contributions from the mismatch components ..............................................................................109
6.8.4.2 Uncertainty contributions: Stage 2: EUT measurement ......................................................................118
6.8.4.2.5 Expanded uncertainty for the receiver Sensitivity measurement...................................................126
6.9 Examples of measurement uncertainty analysis (Stripline)............................................................................126
6.9.1 Introduction...............................................................................................................................................126
6.9.2 Example 1: Verification procedure ...........................................................................................................126
6.9.2.1 Uncertainty contributions: Stage 1: Direct attenuation measurement .................................................127
6.9.2.1.2 Contributions from individual components ...................................................................................130
6.9.2.2 Uncertainty contributions: Stage 2: Radiated attenuation measurement .............................................133
6.9.2.2.2 Contributions from individual components ...................................................................................137
6.9.2.2.5 Expanded uncertainty for the verification procedure.....................................................................141
6.9.3 Example 2: The measurement of a receiver parameter (Sensitivity) ........................................................141
6.9.3.1 Uncertainty contributions: Stage 1: EUT measurement ......................................................................142
6.9.3.2 Uncertainty contributions: Stage 2: Field measurement using the results of the verification
procedure.............................................................................................................................................148
6.9.3.2.1 Expanded uncertainty for the receiver sensitivity measurement ...................................................149
6.9.3.3 Uncertainty contributions: Stage 2: Field measurement using a monopole ........................................149
6.9.3.3.4 Summary table of contributions ....................................................................................................155
6.9.3.4 Uncertainty contributions: Stage 2: Field measurement using 3-axis probe .......................................155
6.10 Uncertainty of fully automated test systems...................................................................................................161
6.10.1 Test system properties ..............................................................................................................................162
6.10.2 General aspects of the measurement uncertainty ......................................................................................162
6.10.3 The "simple" test system...........................................................................................................................163
6.10.3.1 Transmitter measurement....................................................................................................................163
6.10.3.1.1 Error analysis.................................................................................................................................164
6.10.3.1.2 Mismatch uncertainty ....................................................................................................................165
6.10.3.2 Receiver measurements.......................................................................................................................167
6.10.3.2.1 Error analysis.................................................................................................................................168
6.10.3.2.2 Mismatch uncertainty ....................................................................................................................169
6.10.4 The "complex" test system........................................................................................................................170
6.10.4.1 Receiver measurements.......................................................................................................................171
6.10.4.1.1 Error analysis.................................................................................................................................173
6.10.4.1.2 Mismatch uncertainties..................................................................................................................175
6.10.4.2 Transmitter measurements ..................................................................................................................180
6.10.4.2.1 Error analysis.................................................................................................................................181
ETSI
6 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.10.4.2.2 Mismatch uncertainties..................................................................................................................184

6.10.5 Summary...................................................................................................................................................196
6.10.5.1 Typical mismatch example .................................................................................................................197
7 Transmitter measurement examples .....................................................................................................199
7.1 Conducted.......................................................................................................................................................199
7.1.1 Frequency error.........................................................................................................................................199
7.1.2 Carrier power ............................................................................................................................................200
7.1.3 Adjacent channel power............................................................................................................................204
7.1.3.1 Adjacent channel power method 1 (Using an adjacent channel power meter)....................................204
7.1.3.2 Adjacent channel power method 2 (Using a spectrum analyser) ........................................................206
7.1.4 Conducted spurious emissions..................................................................................................................207
7.1.4.1 Direct reading method.........................................................................................................................207
7.1.4.2 Substitution method ............................................................................................................................210
7.1.5 Intermodulation attenuation ......................................................................................................................212
7.1.6 Attack time................................................................................................................................................215
7.1.6.1 Frequency behaviour (attack)..............................................................................................................215
7.1.6.2 Power behaviour (attack) ....................................................................................................................216
7.1.7 Release time..............................................................................................................................................217
7.1.7.1 Frequency behaviour (release) ............................................................................................................217
7.1.7.2 Power behaviour (release)...................................................................................................................217
7.1.8 Transient behaviour of the transmitter ......................................................................................................217
7.1.8.1 Transient frequency behaviour............................................................................................................217
7.1.8.2 Power level slope ................................................................................................................................218
7.1.9 Frequency deviation..................................................................................................................................219
7.1.9.1 Maximum permissible frequency deviation ........................................................................................219
7.1.9.2 Response of the transmitter to modulation frequencies above 3 kHz .................................................220
7.2 Radiated tests..................................................................................................................................................221
7.2.1 Frequency error (30 MHz to 1 000 MHz).................................................................................................221
7.2.1.1 Anechoic Chamber..............................................................................................................................221
7.2.1.1.1 Contributions from the measurement ............................................................................................221
7.2.1.1.2 Expanded uncertainty ....................................................................................................................221
7.2.1.2 Anechoic Chamber with a ground plane .............................................................................................221
7.2.1.3 Open Area Test Site ............................................................................................................................222
7.2.1.4 Stripline...............................................................................................................................................222
7.2.1.5 Test fixture ..........................................................................................................................................222
7.2.2 Effective radiated power (30 MHz to 1 000 MHz) ...................................................................................223
7.2.2.1.1 Uncertainty contributions: Stage one: EUT measurement.............................................................223
7.2.2.1.2 Uncertainty contributions: Stage two: Substitution .......................................................................224
7.2.2.2.2 Uncertainty contributions: Stage two: Substitution measurement.................................................227
7.2.2.4 Stripline...............................................................................................................................................230
7.2.2.5 Test fixture ..........................................................................................................................................230
7.2.3 Radiated spurious emissions .....................................................................................................................231
ETSI
7 ETSI TR 100 028-1 V1.4.1 (2001-12)
7.2.3.1.2 Uncertainty contributions: Stage two: Substitution .......................................................................232

7.2.3.3.3 Expanded uncertainty..........................................................................................................................238
7.2.3.4 Stripline...............................................................................................................................................238
7.2.3.5 Test fixture ..........................................................................................................................................238
7.2.4 Adjacent channel power............................................................................................................................238
7.2.4.4 Stripline...............................................................................................................................................239
7.2.4.5 Test fixture ..........................................................................................................................................239
Annex A: Bibliography ........................................................................................................................240

History ............................................................................................................................................................241
ETSI
8 ETSI TR 100 028-1 V1.4.1 (2001-12)
Intellectual Property Rights

IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (http://webapp.etsi.org/IPR/home.asp).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This Technical Report (TR) has been produced by ETSI Technical Committee Electromagnetic compatibility and Radio
spectrum Matters (ERM).
The present document has been split into two parts, due to practical limitations. However, its has to be considered
altogether; unfortunately, annexes such as D and E which provide theoretical support for the general presentation
provided in the present document, are included in TR 100 028-2 [8].
Version 1.4.1 includes also work on fully automated test systems (see clause 6.10 of the present document).
A presentation has been also added in order to provide a general overview of the approach used in the present document
(see file "MeasurementUncertainties_V141c.ppt") which is available in tr_10002801v010401p0.zip.
Introduction
The present document has been written to clarify the many problems associated with the calculation, interpretation and
application of measurement uncertainty and is expected to be used, in particular, by accredited test laboratories
performing measurements.
In ETR 028 [6] Edition 2, the area of data communication measurement uncertainties has been addressed and added to
the work on analogue measurement uncertainties found in the first edition of the present document; in addition the
diagrams had been standardized and minor editorial corrections had been carried out.
In version 1.3.1, the document has been updated to include a number of enhancements, as a result of work done in the
preparation of TR 102 273 [3] (which covers radiated measurements, now TR 102 273 [3]) and of further work done by
WG RP 02, in particular in the theoretical area (see annex D of TR 100 028-2 [8]). Clause 7 of the present document
and clause 4 of TR 100 028-2 [8] have been considerably revised during year 2000. A set of files (spread sheets)
illustrating the methods proposed has also been added.
The present document is intended to provide, for the relevant standards, methods of calculating the measurement
uncertainty relating to the assessment of the performance of radio equipment. The present document is not intended to
replace any test methods in the relevant standards although clauses 5, 6 and 7 (in the present document) contain brief
descriptions of each measurement (such descriptions are just intended to support the explanations relating to the
evaluation of the uncertainties).
More precisely, the basic purpose of the present document is to:
- provide the method of calculating the total measurement uncertainty (see, in particular annex D of
TR 100 028-2 [8]) and clauses 1 to 5 of the present document);
- provide the maximum acceptable "window" of measurement uncertainty (see table B.1 in TR 100 028-2 [8]),
when calculated using the methods described in the present document;
ETSI
9 ETSI TR 100 028-1 V1.4.1 (2001-12)
- provide the equipment under test dependency functions (see table F.1 (in TR 100 028-2 [8])) which shall be used
in the calculations unless these functions are evaluated by the individual laboratories;
- provide a recommended method of applying the uncertainties in the interpretation of the results (see annex C in
TR 100 028-2 [8]).
Although the present document has been written in a way to cover a larger spread of equipment than what is actually
stated in the scope (in order to help as much as possible) the particular aspects needed regarding some technologies such
as TDMA may have been left out, even though the general approach to measurement uncertainties and the theoretical
background is, in principle, independent of the technology.
Hence, the present document is applicable to measurement methodology in a broad sense but care should be taken when
using it to draft new standards or when applying it to a particular technology such as TDMA or CDMA.
In an attempt to help the user and in order to clarify the particular aspects of each method, a number of examples have
been given (including spread sheets relating to clause 7 of the present document and clause 4 of TR 100 028-2 [8]).
However, these examples may have been drafted by different authors. In a number of cases, simplifications may have
been introduced (e.g. Log (1 + x) = x: simplifications and, hopefully, not real errors), in order to reach practical
conclusions, while avoiding supplementary complications.
As a result, examples covering similar areas may not be fully consistent. The reader is therefore expected to understand
fully the theoretical basis underlying the present document (annex D in TR 100 028-2 [8]) provides the basis for the
theoretical approach) and to exercise his own judgement while using the present document.
As a result, under no circumstances, could ETSI be held for responsible for any consequence of the usage of the present
document.
ETSI
10 ETSI TR 100 028-1 V1.4.1 (2001-12)
1 Scope
The present document provides a method to be applied to all the applicable deliverables, and supports TR 100 027 [2].
It covers the following aspects relating to measurements:
a) methods for the calculation of the total uncertainty for each of the measured parameters;
b) recommended maximum acceptable uncertainties for each of the measured parameters;
c) a method of applying the uncertainties in the interpretation of the results.
The present document provides the methods of evaluating and calculating the measurement uncertainties and the
required corrections on measurement conditions and results (these corrections are necessary in order to remove the
errors caused by certain deviations of the test system due to its known characteristics (such as the RF signal path
attenuation and mismatch loss, etc.)).
2 References
For the purposes of this Technical Report (TR) the following references apply:
[1] Guide to the Expression of Uncertainty in Measurement (International Organization for

Standardization, Geneva, Switzerland, 1995).
[2] ETSI TR 100 027: "Electromagnetic compatibility and Radio spectrum Matters (ERM); Methods
of measurement for private mobile radio equipment".
[3] ETSI TR 102 273 (all parts): "Electromagnetic compatibility and Radio spectrum Matters (ERM);
Improvement of Radiated Methods of Measurement (using test sites) and evaluation of the
corresponding measurement uncertainties".
[4] ITU-T Recommendation O.41: "Psophometer for use on telephone-type circuits".
[5] Void.
[6] ETSI ETR 028: "Radio Equipment and Systems (RES); Uncertainties in the measurement of
mobile radio equipment characteristics".
[7] EN 55020: "Electromagnetic immunity of broadcast receivers and associated equipment".
[8] ETSI TR 100 028-2: "Electromagnetic compatibility and Radio spectrum Matters (ERM);
Uncertainties in the measurement of mobile radio equipment characteristics; Part 2".
3 Definitions, symbols and abbreviations
3.1 Definitions
For the purposes of the present document, the following terms and definitions apply:
accuracy: This term is defined, in relation to the measured value, in clause 4.1.1; it has also been used in the rest of the
document in relation to instruments.
AF load: resistor of sufficient power rating to accept the maximum audio output power from the EUT
NOTE: The value of the resistor should be that stated by the manufacturer and should be the impedance of the
audio transducer at 1 000 Hz.
In some cases it may be necessary to place an isolating transformer between the output terminals of the
receiver under test and the load.
ETSI
11 ETSI TR 100 028-1 V1.4.1 (2001-12)
AF termination: any connection other than the audio frequency load which may be required for the purpose of testing
the receiver (i.e. in a case where it is required that the bit stream be measured, the connection may be made, via a
suitable interface, to the discriminator of the receiver under test)
NOTE: The termination device should be agreed between the manufacturer and the testing authority and details
should be included in the test report. If special equipment is required then it should be provided by the
manufacturer.
antenna: part of a transmitting or receiving system that is designed to radiate or to receive electromagnetic waves
antenna factor: quantity relating the strength of the field in which the antenna is immersed to the output voltage across
the load connected to the antenna
NOTE: When properly applied to the meter reading of the measuring instrument, yields the electric field strength
in V/m or the magnetic field strength in A/m.
antenna gain: ratio of the maximum radiation intensity from an (assumed lossless) antenna to the radiation intensity
that would be obtained if the same power were radiated isotropically by a similarly lossless antenna
bit error ratio: ratio of the number of bits in error to the total number of bits
combining network: network allowing the addition of two or more test signals produced by different sources (e.g. for
connection to a receiver input)
NOTE: Sources of test signals should be connected in such a way that the impedance presented to the receiver
should be 5O Ω. The effects of any intermodulation products and noise produced in the signal generators
should be negligible.
correction factor: numerical factor by which the uncorrected result of a measurement is multiplied to compensate for
an assumed systematic error
confidence level: probability of the accumulated error of a measurement being within the stated range of uncertainty of
measurement
directivity: ratio of the maximum radiation intensity in a given direction from the antenna to the radiation intensity
averaged over all directions (i.e. directivity = antenna gain + losses)
duplex filter: duplex filter is a device fitted internally or externally to a transmitter/receiver combination to allow
simultaneous transmission and reception with a single antenna connection
error of measurement (absolute): result of a measurement minus the true value of the measurand
error (relative): ratio of an error to the true value
estimated standard deviation: from a sample of n results of a measurement the estimated standard deviation is given
by the formula:
∑(x − x )
i =1
i
2
σ=
n −1
xi being the ith result of measurement (i = 1, 2, 3, ..., n) and x the arithmetic mean of the n results considered.
A practical form of this formula is:
X2
Y−
σ = n
n −1
Where X is the sum of the measured values and Y is the sum of the squares of the measured values.
The term standard deviation has also been used in the present document to characterize a particular probability
density. Under such conditions, the term standard deviation may relate to situations where there is only one result for a
measurement.
ETSI
12 ETSI TR 100 028-1 V1.4.1 (2001-12)
expansion factor: multiplicative factor used to change the confidence level associated with a particular value of a
measurement uncertainty
The mathematical definition of the expansion factor can be found in clause D.5.6.2.2.
extreme test conditions: extreme test conditions are defined in terms of temperature and supply voltage
NOTE: Tests should be made with the extremes of temperature and voltage applied simultaneously. The upper
and lower temperature limits are specified in the relevant ETS. The test report should state the actual
temperatures measured
error (of a measuring instrument): indication of a measuring instrument minus the (conventional) true value
free field: field (wave or potential) which has a constant ratio between the electric and magnetic field intensities
free space: region free of obstructions and characterized by the constitutive parameters of a vacuum
impedance: measure of the complex resistive and reactive attributes of a component in an alternating current circuit
impedance (wave): complex factor relating the transverse component of the electric field to the transverse component
of the magnetic field at every point in any specified plane, for a given mode
influence quantity: quantity which is not the subject of the measurement but which influences the value of the quantity
to be measured or the indications of the measuring instrument
intermittent operation: manufacturer should state the maximum time that the equipment is intended to transmit and
the necessary standby period before repeating a transmit period
isotropic radiator: hypothetical, lossless antenna having equal radiation intensity in all directions
limited frequency range: specified smaller frequency range within the full frequency range over which the
measurement is made
NOTE: The details of the calculation of the limited frequency range should be given in the relevant deliverable.
maximum permissible frequency deviation: maximum value of frequency deviation stated for the relevant channel
separation in the relevant deliverable
measuring system: complete set of measuring instruments and other equipment assembled to carry out a specified
measurement task
measurement repeatability: Closeness of the agreement between the results of successive measurements of the same
measurand carried out subject to all the following conditions:
- the same method of measurement;
- the same observer;
- the same measuring instrument;
- the same location;
- the same conditions of use;
- repetition over a short period of time
measurement reproducibility: Closeness of agreement between the results of measurements of the same measurand,
where the individual measurements are carried out changing conditions such as:
- method of measurement;
- observer;
- measuring instrument;
- location;
ETSI
13 ETSI TR 100 028-1 V1.4.1 (2001-12)
- conditions of use;
- time.
measurand: quantity subjected to measurement
noise gradient of EUT: function characterizing the relationship between the RF input signal level and the performance
of the EUT, e.g. the SINAD of the AF output signal
nominal frequency: defined as one of the channel frequencies on which the equipment is designed to operate
nominal mains voltage: declared voltage or any of the declared voltages for which the equipment was designed
normal test conditions: defined in terms of temperature, humidity and supply voltage stated in the relevant deliverable
normal deviation: frequency deviation for analogue signals which is equal to 12 % of the channel separation
psophometric weighting network: Should be as described in ITU-T Recommendation O.41
polarization: for an electromagnetic wave, this is the figure traced as a function of time by the extremity of the electric
vector at a fixed point in space
quantity (measurable): attribute of a phenomenon or a body which may be distinguished qualitatively and determined
quantitatively
rated audio output power: maximum output power under normal test conditions, and at standard test modulations, as
declared by the manufacturer
rated radio frequency output power: maximum carrier power under normal test conditions, as declared by the
manufacturer
shielded enclosure: structure that protects its interior from the effects of an exterior electric or magnetic field, or
conversely, protects the surrounding environment from the effect of an interior electric or magnetic field
SINAD sensitivity: minimum standard modulated carrier-signal input required to produce a specified SINAD ratio at
the receiver output
stochastic (random) variable: variable whose value is not exactly known, but is characterized by a distribution or
probability function, or a mean value and a standard deviation (e.g. a measurand and the related measurement
uncertainty)
test load: 50 Ω substantially non-reactive, non-radiating power attenuator which is capable of safely dissipating the power
from the transmitter
test modulation: test modulating signal is a baseband signal which modulates a carrier and is dependent upon the type of
EUT and also the measurement to be performed
trigger device: circuit or mechanism to trigger the oscilloscope timebase at the required instant
NOTE: It may control the transmit function or inversely receive an appropriate command from the transmitter.
uncertainty: parameter, associated with the result of a measurement, that characterizes the dispersion of the values that
could reasonably be attributed to that measurement
uncertainty (random): component of the uncertainty of measurement which, in the course of a number of
measurements of the same measurand, varies in an unpredictable way (and has not being considered otherwise)
uncertainty (systematic): component of the uncertainty of measurement which, in the course of a number of
measurements of the same measurand remains constant or varies in a predictable way
uncertainty (type A): uncertainties evaluated using the statistical analysis of a series of observations
uncertainty (type B): uncertainties evaluated using other means than the statistical analysis of a series of observations
ETSI
14 ETSI TR 100 028-1 V1.4.1 (2001-12)
uncertainty (limits of uncertainty of a measuring instrument): extreme values of uncertainty permitted by

specifications, regulations etc. for a given measuring instrument
NOTE: This term is also known as "tolerance".
uncertainty (standard): for each individual uncertainty component, an expression characterizing the uncertainty for
that component
NOTE: It is the standard deviation of the corresponding distribution.
uncertainty (combined standard): uncertainty characterizing the complete measurement or part thereof
NOTE: It is calculated by combining appropriately the standard uncertainties for each of the individual
contributions identified in the measurement considered or in the part of it which has been considered.
In the case of additive components (linearly combined components where all the corresponding
coefficients are equal to one) and when all these contributions are independent of each other (stochastic),
this combination is calculated by using the Root of the Sum of the Squares (the RSS method). A more
complete methodology for the calculation of the combined standard uncertainty is given in annex D;
see in particular, clause D.3.12 of TR 100 028-2 [8].
uncertainty (expanded): expanded uncertainty is the uncertainty value corresponding to a specific confidence level
different from that inherent to the calculations made in order to find the combined standard uncertainty
NOTE: The combined standard uncertainty is multiplied by a constant to obtain the expanded uncertainty limits
(see clause 5.3 of the present document and also clause D.5 (and more specifically clause D.5.6.2 of
TR 100 028-2 [8]).
upper specified AF limit: upper specified audio frequency limit is the maximum audio frequency of the audio pass-band
and is dependent on the channel separation
wanted signal level: for conducted measurements the wanted signal level is defined as a level of +6 dB/µV emf referred
to the receiver input under normal test conditions. Under extreme test conditions the value is +12 dB/µV emf
NOTE: For analogue measurements the wanted signal level has been chosen to be equal to the limit value of the
measured usable sensitivity. For bit stream and message measurements the wanted signal has been chosen
to be +3 dB above the limit value of measured usable sensitivity.
3.2 Symbols
For the purposes of the present document, the following symbols apply:
β 2π/λ (radians/m)
γ incidence angle with ground plane (°)
λ wavelength (m)
φH phase angle of reflection coefficient (°)
η 120π Ω - the intrinsic impedance of free space (Ω)
µ permeability (H/m)
AFR antenna factor of the receive antenna (dB/m)
AFT antenna factor of the transmit antenna (dB/m)
AFTOT mutual coupling correction factor (dB)
Ccross cross correlation coefficient
D(θ,φ) directivity of the source
d distance between dipoles (m)
δ skin depth (m)
d1 an antenna or EUT aperture size (m)
d2 an antenna or EUT aperture size (m)
ddir path length of the direct signal (m)
drefl path length of the reflected signal (m)
E electric field intensity (V/m)
EDHmax calculated maximum electric field strength in the receiving antenna height scan from a half
wavelength dipole with 1 pW of radiated power (for horizontal polarization) (µV/m)
ETSI
15 ETSI TR 100 028-1 V1.4.1 (2001-12)
EDVmax calculated maximum electric field strength in the receiving antenna height scan from a half
wavelength dipole with 1 pW of radiated power (for vertical polarization) (µV/m)
eff antenna efficiency factor
φ angle (°)
∆f bandwidth (Hz)
f frequency (Hz)
G(θ,φ) gain of the source (which is the source directivity multiplied by the antenna efficiency factor)
H magnetic field intensity (A/m)
I0 the (assumed constant) current (A)
Im the maximum current amplitude
k 2π/λ
k a factor from Student's distribution
k Boltzmann's constant (1,38 x 10 - 23 J/°K)
K relative dielectric constant
l the length of the infinitesimal dipole (m)
L the overall length of the dipole (m)
l the point on the dipole being considered (m)
λ wavelength (m)
Pe (n) probability of error n
Pp (n) probability of position n
Pr antenna noise power (W)
Prec power received (W)
Pt power transmitted (W)
θ angle (°)
ρ reflection coefficient
r the distance to the field point (m)
ρg reflection coefficient of the generator part of a connection
ρl reflection coefficient of the load part of the connection
Rs equivalent surface resistance (Ω)
σ conductivity (S/m)
σ standard deviation
SNRb* Signal to noise ratio at a specific BER
SNRb Signal to noise ratio per bit
TA antenna temperature (°K)
U the expanded uncertainty corresponding to a confidence level of x %: U = k × uc
uc the combined standard uncertainty
ui general type A standard uncertainty
ui01 random uncertainty
uj general type B uncertainty
uj01 reflectivity of absorbing material: EUT to the test antenna
uj02 reflectivity of absorbing material: substitution or measuring antenna to the test antenna
uj03 reflectivity of absorbing material: transmitting antenna to the receiving antenna
uj04 mutual coupling: EUT to its images in the absorbing material
uj05 mutual coupling: de-tuning effect of the absorbing material on the EUT
uj06 mutual coupling: substitution, measuring or test antenna to its image in the absorbing material
uj07 mutual coupling: transmitting or receiving antenna to its image in the absorbing material
uj08 mutual coupling: amplitude effect of the test antenna on the EUT
uj09 mutual coupling: de-tuning effect of the test antenna on the EUT
uj10 mutual coupling: transmitting antenna to the receiving antenna
uj11 mutual coupling: substitution or measuring antenna to the test antenna
uj12 mutual coupling: interpolation of mutual coupling and mismatch loss correction factors
uj13 mutual coupling: EUT to its image in the ground plane
uj14 mutual coupling: substitution, measuring or test antenna to its image in the ground plane
uj15 mutual coupling: transmitting or receiving antenna to its image in the ground plane
ETSI
16 ETSI TR 100 028-1 V1.4.1 (2001-12)
uj16 range length

uj17 correction: off boresight angle in the elevation plane
uj18 correction: measurement distance
uj19 cable factor
uj20 position of the phase centre: within the EUT volume
uj21 positioning of the phase centre: within the EUT over the axis of rotation of the turntable
uj22 position of the phase centre: measuring, substitution, receiving, transmitting or test antenna
uj23 position of the phase centre: LPDA
uj24 stripline: mutual coupling of the EUT to its images in the plates
uj25 stripline: mutual coupling of the 3-axis probe to its image in the plates
uj26 stripline: characteristic impedance
uj27 stripline: non-planar nature of the field distribution
uj28 stripline: field strength measurement as determined by the 3-axis probe
uj29 stripline: Transform Factor
uj30 stripline: interpolation of values for the Transform Factor
uj31 stripline: antenna factor of the monopole
uj32 stripline: correction factor for the size of the EUT
uj33 stripline: influence of site effects
uj34 ambient effect
uj35 mismatch: direct attenuation measurement
uj36 mismatch: transmitting part
uj37 mismatch: receiving part
uj38 signal generator: absolute output level
uj39 signal generator: output level stability
uj40 insertion loss: attenuator
uj41 insertion loss: cable
uj42 insertion loss: adapter
uj43 insertion loss: antenna balun
uj44 antenna: antenna factor of the transmitting, receiving or measuring antenna
uj45 antenna: gain of the test or substitution antenna
uj46 antenna: tuning
uj47 receiving device: absolute level
uj48 receiving device: linearity
uj49 receiving device: power measuring receiver
uj50 EUT: influence of the ambient temperature on the ERP of the carrier
uj51 EUT: influence of the ambient temperature on the spurious emission level
uj52 EUT: degradation measurement
uj53 EUT: influence of setting the power supply on the ERP of the carrier
uj54 EUT: influence of setting the power supply on the spurious emission level
uj55 EUT: mutual coupling to the power leads
uj56 frequency counter: absolute reading
uj57 frequency counter: estimating the average reading
uj58 Salty-man/Salty-lite: human simulation
uj59 Salty-man/Salty-lite: field enhancement and de-tuning of the EUT
uj60 Test Fixture: effect on the EUT
uj61 Test Fixture: climatic facility effect on the EUT
Vdirect received voltage for cables connected via an adapter (dBµV/m)
Vsite received voltage for cables connected to the antennas (dBµV/m)
W0 radiated power density (W/m2)
Other symbols which are used only in annexes D or E of TR 100 028-2 [8] are defined in the corresponding annexes.
ETSI
17 ETSI TR 100 028-1 V1.4.1 (2001-12)
3.3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
AF Audio Frequency
BER Bit Error Ratio
BIPM International Bureau of Weights and Measures (Bureau International des Poids et Mesures)
c calculated on the basis of given and measured data
d derived from a measuring equipment specification
emf electromotive force
EUT Equipment Under Test
FSK Frequency Shift Keying
GMSK Gaussian Minimum Shift Keying
GSM Global System for Mobile telecommunication (Pan European digital telecommunication system)
m measured
NSA Normalized Site Attenuation
p power level value
v voltage level value
r indicates rectangular distribution
RF Radio Frequency
RSS Root-Sum-of-the-Squares
u indicates U-distribution
VSWR Voltage Standing Wave Ratio
4 Introduction to measurement uncertainty

This clause gives the general background to the subject of measurement uncertainty and is also the basis of TR 102 273
[3]. It covers methods of evaluating both individual components and overall system uncertainties and ends with a
discussion of the generally accepted present day approach to the calculation of overall measurement uncertainty.
For further details and for the basis of a theoretical approach, please see annex D of TR 100 028-2 [8].
An outline of the extensions and improvements recommended is also included in this clause.
This clause should be viewed as introductory material for clauses 5 and 6, and to some extent, also for annex D of
TR 100 028-2 [8].
4.1 Background to measurement uncertainty

4.1.1 Commonly used terms
UNCERTAINTY is that part of the expression of the result of a measurement which states the range of values within
which the true value is estimated to lie.
ACCURACY is an estimate of the closeness of the measured value to the true value. An accurate measurement is one in
which the uncertainties are small. This term is not to be confused with the terms PRECISION or REPEATABILITY
which characterize the ability of a measuring system to give identical indications or responses for repeated applications
of the same input quantity.
Measuring exactly a quantity (referred to as the measurand) is an ideal which cannot be attained in practical
measurements. In every measurement a difference exists between the TRUE VALUE and the MEASURED VALUE.
This difference is termed "THE ABSOLUTE ERROR OF THE MEASUREMENT". This error is defined as follows:
Absolute error = the measured value - the true value.
Since the true value is never known exactly, it follows that the absolute error cannot be known exactly either. The above
formula is the defining statement for the terms of ABSOLUTE ERROR and TRUE VALUE, but, as a result of neither
ever being known, it is recommended that these terms are never used.
ETSI
18 ETSI TR 100 028-1 V1.4.1 (2001-12)
In practice, many aspects of a measurement can be controlled (e.g. temperature, supply voltage, signal generator output
level, etc.) and by analysing a particular measurement set-up, the overall uncertainty can be assessed, thereby providing
upper and lower UNCERTAINTY BOUNDS within which the true value is believed to lie.
The overall uncertainty of a measurement is an expression of the fact that the measured value is only one of an infinite
number of possible values dispersed (spread) about the true value.
This is further developed in clause D.5.6 of TR 100 028-2 [8].
4.1.2 Assessment of upper and lower uncertainty bounds

One method of providing upper and lower bounds is by straightforward arithmetic calculation in the worst case
condition, using the individual uncertainty contributions. This method can be used to arrive at a value each side of the
measured result within which, there is utmost confidence (100 %) that the true value lies (see also clause D.5.6.1 in
TR 100 028-2 [8]).
When estimating the measurement uncertainty in the worst case e.g. by simply adding the uncertainty bounds (in
additive situations), (extremely) pessimistic uncertainty bounds are often found. This is because the case when all the
individual uncertainty components act to their maximum effect in the same direction at the same time is, in practice,
very unlikely to happen (it has to be noted, however, that the usage of expansion factors in order to increase the
confidence levels (see also clause 5.3.1 and clauses D.5.6.2.2 and D.3.3.5.2 in TR 100 028-2 [8]) may have a balancing
effect).
To overcome this (very) pessimistic calculation of the lower and upper bounds, a more realistic approach to the
calculation of overall uncertainty needs to be taken (i.e. a probabilistic approach).
The method presented in the present document is based on the approach to expressing uncertainty in measurement as
recommended by the Comité International des Poids et Mesures (CIPM) in 1981. This approach is founded on
Recommendation INC-1 (1980) of the Working Group on the Statement of Uncertainties. This group was convened in
1980 by the Bureau International des Poids et Mesures (BIPM) as a consequence of a request by the Comité that the
Bureau study the question of reaching an international consensus on expressing uncertainty in measurement.
Recommendation INC-1 (1980) led to the development of the Guide to the Expression of Uncertainty in
Measurement [1] (the Guide), which was prepared by the International Organization for Standardization Technical
Advisory Group 4 (ISOTAG 4), Working Group 3. The Guide was the most complete reference on the general
application of the BIPM approach to expressing measurement uncertainty. Further theoretical analysis has been
introduced in the third edition of the present document (see, in particular, annexes D and E in TR 100 028-2 [8]).
Although the Guide represented the current international view of how to express uncertainty it is a rather lengthy
document that is not easily interpreted for radiated measurements. The guidance given in the present document is
intended to be applicable to radio measurements but since the Guide itself is intended to be generally applicable to
measurement results, it should be consulted for additional details, if needed.
The method in both the present document and the Guide apply statistical/probabilistic analysis to estimate the overall
uncertainties of a measurement and to provide associated confidence levels. They depend on knowing the magnitude
and distribution of the individual uncertainty components. This approach is commonly known as the BIPM method.
Basic to the BIPM method is the representation of each individual uncertainty component that contributes to the overall
measurement uncertainty by an estimated standard deviation, termed standard uncertainty [1], with suggested symbol
u.
All individual uncertainties are categorized as either type A or type B. Type A uncertainties, symbol ui, are estimated by
statistical methods applied to repeated measurements, whilst type B uncertainties, symbol uj, are estimated by means of
available information and experience.
The combined standard uncertainty [1], symbol uc, of a measurement is calculated by combining the standard
uncertainties for each of the individual contributions identified. In the case where the underlying physical effects are
additive, this is done by applying the "Root of the Sum of the Squares (the RSS)" method (see also clause D.3.3 in
TR 100 028-2 [8]) under the assumption that all contributions are stochastic i.e. independent of each other.
The table included in clause D.3.12 of TR 100 028-2 [8] provides the way in which should be handled contributions to
the uncertainty which correspond to physical effects which are not additive. Clause D.5 of the same annex provides an
overview of several general methods.
ETSI
19 ETSI TR 100 028-1 V1.4.1 (2001-12)
The resulting combined standard uncertainty can then be multiplied by a constant kxx to give the uncertainty limits
(bounds), termed expanded uncertainty [1], in order to provide a confidence level of xx %. This is further discussed in
clause D.5.6.2 of TR 100 028-2 [8].
One of the main assumptions when calculating uncertainty using the basic BIPM method is that the combined standard
uncertainty of a measurement has a Normal or Gaussian distribution (see also clause D.1.3.4 in TR 100 028-2 [8]) with
an associated standard deviation (the present document often uses the term Normal). This may be true when there is an
infinite number of contributions in the uncertainty, which is generally not the case in the examples discussed in the
present document (an interesting example is provided in clause D.3.3.5.2.2 of TR 100 028-2 [8]).
Should the combined standard uncertainty correspond to a Normal distribution, then the multiplication by the
appropriate constant (expansion factor) will provide the sought confidence level.
The case where the combined standard uncertainty corresponds to non-Gaussian distributions is also considered in
clauses D.5.6.2.3 and D.5.6.2.4 of TR 100 028-2 [8].
The Guide defines the combined standard uncertainty for this distribution uc, as equal to the standard deviation of a
corresponding Normal distribution. The mean value is assumed to be zero as the measured result is corrected for all
known errors. Based on this assumption, the uncertainty bounds corresponding to any confidence level can be
calculated as kxx × uc (see also clause D.5.6.2 of TR 100 028-2 [8]).
To illustrate the true meaning of a typical final statement of measurement uncertainty using this method, if the
combined standard uncertainty is associated with a Normal distribution, confidence levels can be assigned as follows:
- 68,3 % confidence level that the true value is within bounds of 1 × uc;
- 95 % confidence within ±1,96 × uc, etc.
Care must be taken in the judgement of which unit is chosen for the calculation of the uncertainty bounds. In some
types of measurements the correct unit is logarithmic (dB); in other measurements it is linear (i.e. V or %). The choice
depends on the model and architecture of the test system. In any measurement there may be a combination of different
types of unit. The present document breaks new ground by giving methods for conversion between units (e.g. dB into
V %, power % into dB, etc.) thereby allowing all types of uncertainty to be combined. Details of the conversion
schemes are given in clause 5, and theoretical support in annexes D and E of TR 100 028-2 [8].
4.1.3 Combination of rectangular distributions

The following example shows that the overall uncertainty, when all contributions of a measurement have the same
rectangular distribution, approaches a Normal distribution.
The case of a discrete approach to a rectangularly distributed function, (the outcome of throwing a die), is shown and
how, with up to 6 individual events simultaneously, (6 dice thrown at the same time) the events combine together to
produce an output increasingly approximating a Normal distribution.
Initially with 1 die the output mean is 3,5 with a rectangularly distributed "error" of ±2,5. With 2 dice the output is 7 ± 5
and is triangularly distributed see figure 1.
ETSI
20 ETSI TR 100 028-1 V1.4.1 (2001-12)
Probability of a number (1 die) Probability of a number (2 dice)
1 2 3 4 5 6 2 3 4 5 6 7 8 9 10 11 12
Figure 1: One and two die outcomes
By increasing the number of dice further through 3, 4, 5 and 6 dice it can be seen from figures 2 and 3, that there is a
central value (most probable outcome) respectively for 2, 3, 4, 5 and 6 dice of (7), (10,5), (14), (17,5) and (21) and an
associated spread of the results that increasingly approximates a Normal distribution. It is possible to calculate the mean
and standard deviation for these events.
Probability of a number (3 dice) Probability of a number (4 dice)
3 4 5 6 7 8 9 10 11 12 1314 15 1617 18 4 6 8 10 12 14 16 18 20 22 24
Figure 2: Three and four die outcomes
Probability of a number (5 dice) Probability of a number (6 dice)
6 8 10 12 14 16 18 20 22 24 26 28 30 10 15 20 25 30 35
Figure 3: Five and six die outcomes
ETSI
21 ETSI TR 100 028-1 V1.4.1 (2001-12)
The practical interpretation of the standard deviation of a Normally distributed quantity is that 68,3 % of all its possible
values will lie within ±1 standard deviation of the mean value, 95,45 % will lie within ±2 standard deviations. Another
way to regard these standard deviations is "as confidence levels", e.g. a confidence level of 68,3 % attaches to one
standard deviation, 95,45 % to two standard deviations.
Using the mathematical definition of a Gaussian (see annex D in TR 100 028-2 [8]), it is possible to calculate the
expanded measurement uncertainty for other confidence levels.
This illustration shows that in the case of individual throws of a die (which corresponds to a set of identical rectangular
distributions since any of the values 1 to 6 is equally likely) the overall probability curve approximates closer and closer
that of a Normal distribution as more dice are used.
The BIPM method extends this principle by combining the individual standard uncertainties to derive a combined
standard uncertainty. The standard uncertainties (corresponding to the distributions of the individual uncertainties) are
all that need to be known (or assumed) to apply this approach. From the assumption that the final combined standard
uncertainty corresponds to a Normal distribution, it is possible to calculate the expanded uncertainty for a given
confidence level.
The confidence level should always be stated in any test report, in the case where the resulting distribution is Gaussian.
In such case, it makes it possible for the user of the measured results to calculate expanded uncertainty figures
corresponding to other confidence levels.
For similar reasons, in the case where there is no evidence that the distribution corresponding to the combined
uncertainty is Normal, the expansion factor, kxx, should be stated in the test report, instead. Usually, for the reasons
stated above, kxx = 1,96 is used (see also clause D.5.6.2 of TR 100 028-2 [8]).
An expansion factor, kxx = 2,00 could also be acceptable; it would provide a confidence level of 95,45 % should the
corresponding distribution be Normal.
4.1.4 Main contributors to uncertainty

The main contributors to the overall uncertainty of a measurement comprise:
- systematic uncertainties: those uncertainties inherent in the test equipment used (instruments, attenuators, cables,
amplifiers, etc.), and in the method employed. These uncertainties cannot always be eliminated (calculated out)
although they may be constant values, however they can often be reduced;
- uncertainties relating to influence quantities i.e. those uncertainties whose magnitudes are dependent on a
particular parameter or function of the EUT. The magnitude of the uncertainty contribution can be calculated, for
example, from the slope of "dB RF level" to "dB SINAD" curve for a receiver or from the slope of a power
supply voltage effect on the variation of a carrier output power or frequency;
- random uncertainties: those uncertainties due to chance events which, on average, are as likely to occur as not to
occur and are generally outside the engineer's control.
NOTE: When making a measurement care must be taken to ensure that the measured value is not affected by
unwanted or unknown influences. Extraneous influences (e.g. ambient signals on an Open Area Test Site)
should be eliminated or minimized by, for example, the use of screened cables.
4.1.5 Other contributors

Other contributors to the overall uncertainty of a measurement can relate to the standard itself:
- the type of measurement (direct field, substitution or conducted) and the test method have an effect on the
uncertainty. These can be the most difficult uncertainty components to evaluate. As an illustration, if the same
measurand is determined by the same method in different laboratories (as in a round robin) or alternatively by
different methods either in the same laboratory or in different laboratories, the results of the testing will often be
widely spread, thereby showing the potential uncertainties of the different measurement types and test methods;
ETSI
22 ETSI TR 100 028-1 V1.4.1 (2001-12)
- a direct field measurement involves only a single testing stage in which the required parameter (ERP, sensitivity,
etc.) is indirectly determined as the received level on a receiving device, or as the output level of a signal
generator, etc., and is subsequently converted to ERP, field strength, etc., by a calculation involving knowledge
of antenna gain, measurement distance, etc. This method, whilst being of short time duration, offers no way of
allowing for imperfections (reflections, mutual coupling effects, etc.) in the test site and can results in large
overall uncertainty values;
- the substitution technique, on the other hand, is a two stage measurement in which the unknown performance of
an EUT (measured in one stage) is directly compared with the "known" performance of some standard (usually
an antenna) in the other stage. This technique therefore subjects both the EUT and the known standard to
(hopefully) the same external influences of reflections, mutual coupling, etc., whose effects on the different
devices are regarded as identical. As a consequence, these site effects are deemed to cancel out (this has also
been addressed in clause D.5.3.2 in TR 100 028-2 [8]). Some residual effects do remain however, (due to
different elevation beamwidths, etc.) but these tend to be small compared to the uncertainties in the direct field
method. All the test methods in the present document are substitution measurements;
- for their part, test methods can contain imprecise and ambiguous instructions which could be open to different
interpretations;
- an inadequate description of the measurand can itself be a source of uncertainty in a measurement. In practice a
measurand cannot be completely described without an infinite amount of information. Because this definition is
incomplete it therefore introduces into the measurement result a component of uncertainty that may or may not
be significant relative to the overall uncertainty required of the measurement. The definition of the measurand
may, for example, be incomplete because:
• it does not specify parameters that may have been assumed, unjustifiably, to have negligible effect
(i.e. coupling to the ground plane, reflections from absorbers or that reference conditions remain
constant);
• it leaves many other matters in doubt that might conceivably affect the measurement (i.e. supply
voltages, the layout of power, signal and antenna cables);
• it may imply conditions that can never be fully met and whose imperfect realization is difficult to take
into account (i.e. an infinite, perfectly conducting ground plane, a free space environment) etc.
Maximum acceptable uncertainties and confidence levels (or expansion factors) are both defined in most ETSI
standards.
4.2 Evaluation of individual uncertainty components

As discussed in clause 4.1.4, uncertainty components can be categorized either as "random" or "systematic". Such
categorization of components of uncertainty can be ambiguous if they are applied too rigorously. For example, a
"random" component of uncertainty in one measurement may become a "systematic" component of uncertainty in
another measurement e.g. where the result of a first measurement is used as a component of a second measurement.
Categorizing the methods of evaluating the uncertainty components rather than the components themselves avoids this
ambiguity.
Instead of "systematic" and "random" uncertainty the types of uncertainty contribution are grouped into two categories:
- type A: those which are evaluated by statistical methods;
- type B: those which are evaluated by other means.
The classification into type A and type B is not meant to indicate that there is any difference in the nature of the
components, it is simply a division based on their means of evaluation. Both types will possess probability distributions
(although they may be governed by different rules), and the uncertainty components resulting from either type may be
quantified by standard deviations.
ETSI
23 ETSI TR 100 028-1 V1.4.1 (2001-12)
4.2.1 Evaluation of type A uncertainties

When we carry out a measurement more than once and find the results are different, the following questions arise:
- What to do with the results?
- How much variation is acceptable?
- When do we suspect the measuring system is faulty?
- Are the conditions repeatable?
Variations in these repeated measurements are assumed to be due to influence and random quantities that affect the
measurement result and cannot be held completely constant. Therefore none of the results is necessarily correct. In
practice, repeated measurements of the same measurand can help us evaluate these type A uncertainties. By treating the
results statistically, we can derive the mean (the best approximation to the "true value") and standard deviation values.
The standard deviation can then be incorporated as a standard uncertainty into the calculation of combined standard
uncertainty, when the corresponding component is part of some measurement system.
Uncertainties determined from repeated measurements are often thought of as statistically rigorous and therefore
absolutely correct. This implies, sometimes wrongly, that their evaluation does not require the application of some
judgement. For example:
- When carrying out a series of measurements do the results represent completely independent repetitions or are
they in some way biased?
- Are we trying to assess the randomness of the measurement system, or the randomness in an individual EUT, or
the randomness in all of the EUT produced?
- Are the means and standard deviations constant, or is there perhaps a drift in the value of an unmeasured
influence quantity during the period of repeated measurements?
- Are the results stable with ambient conditions?
If all of the measurements are on a single EUT, whereas the requirement is for sampling, then the observations have not
been independently repeated. An estimate of the standard uncertainty arising from possible differences among
production EUT should, in this case, be incorporated into the combined standard uncertainty calculation along with the
calculated standard uncertainty of the repeated observations made on the single equipment (e.g. for characterizing a set
of pieces of equipment).
If an instrument is calibrated against an internal reference as part of the measurement procedure, (such as the "cal out"
reference on a spectrum analyser), then the calibration should be carried out as part of every repetition, even if it is
known that the drift is small during the period in which observations are made.
If the EUT is rotated during a radiated test on a test site and the azimuth angle read, it should be rotated and read for
each repetition of the measurement, for there may be a variation both in received level and in azimuth reading, even if
everything else is constant.
If a number of measurements have been carried out on the same EUT/types of EUT, but in two groups spaced apart in
time, the arithmetic means of the results of the first and second groups of measurements and their experimentally
derived means and standard deviations may be calculated and compared. This will enable a judgement to be made as to
whether any time varying effects are statistically significant.
4.2.2 Evaluation of type B uncertainties

Some examples of type B uncertainties are:
- mismatch;
- losses in cables and components;
- non linearities in instruments;
- antenna factors.
ETSI
24 ETSI TR 100 028-1 V1.4.1 (2001-12)
Type B uncertainties do not reveal themselves as fluctuations as do type A uncertainties; they can only be assessed by
careful analysis of test and calibration data.
For incorporation into an overall analysis, the magnitudes and distributions of type B uncertainties can be estimated
based on:
- manufacturers' information/specification about instruments and components in the test set-up;
- data in calibration certificates (if the history of the instrument is known);
- experience with the behaviour of the instruments.
4.2.3 Uncertainties relating to influence quantities

Uncertainties relating to influence quantities are, as a result of the way they are treated in the present document,
regarded as a subgroup of type B uncertainties. Some examples of influence quantities are:
- power supply;
- ambient temperature;
- time/duty cycle.
Their effect is evaluated using some relationship between the measured parameter e.g. output power and the influence
quantity e.g. supply voltage.
Dependency functions (e.g. the relationship between output power and the fluctuating quantity), as those given in the
present document, should be used to calculate the properties corresponding to the effect considered.
A theoretical approach to influence quantities and dependency functions can be found in TR 100 028-2 [8] (see
clause D.4).
4.3 Methods of evaluation of overall measurement uncertainty

The uncertainty of the measurement is a combination of many components.
Some of these components may be evaluated from the statistical distributions of the results of a series of measurements
(type A uncertainty) whilst other components are evaluated from assumed probability distributions based on experience
or other information (type B uncertainty).
The exact error of a result of a measurement is, in general, unknown and unknowable. All that can be done is to
estimate the values of all quantities likely to contribute to the combined standard uncertainty, including those
uncertainties associated with corrections for recognized systematic offset effects. With knowledge of the magnitudes of
their individual standard uncertainties, it is then possible to calculate the combined standard uncertainty of the
measurement.
At present the assessment of the number of uncertainty components for any particular test is very variable. Whilst some
general agreement has been reached on the manner in which individual uncertainties should be combined (the BIPM
method, see also the discussion of such methods in TR 100 028-2 [8], annex D, in particular, in clause D.5), no such
agreement has been arrived at concerning the identity of those individual components. Consequently, it is left to the
particular test house/engineer/etc. to decide the contributory uncertainties, and to assess which are independent and
which are not. This can lead to considerable test house to test house variation for the same test and is heavily dependent,
in general, on the experience of the test engineer.
A model of the measurement can assist in the evaluation of combined standard uncertainty since it will enable all
known individual components of uncertainty to be rigorously included in the analysis, and correctly combined (see
annex D in TR 100 028-2 [8], and, in particular, the table in clause D.3.12).
ETSI
25 ETSI TR 100 028-1 V1.4.1 (2001-12)
4.4 Summary
The measured result can be affected by many variables, some of which are shown in figure 4.
Influence quantities Random Inadequate definition

Temperature, supply voltage etc. uncertainties of the measurand
Measuring
Equipment system
Systematic Measured
under Coupling
uncertainties result
test
Statistical
Contributions from Exisiting fluctuations
the test method knowledge Corrections
Figure 4: The measurement model
4.5 Overview of the approach of the present document

The present document proposes an approach to the calculation of the combined standard uncertainty of a measurement
which includes solutions to the present day imperfections.
For example, in clause 5, a technique is put forward for converting linear standard deviations into logarithmic ones (and
vice versa) so that all uncertainty contributions for a particular test can be combined in the same units (dB, Voltage %
or power %), and as stated above, comprehensive lists of the individual uncertainty sources for the tests are attached.
Instructions within the test methods have been made more detailed and thereby less ambiguous.
A global approach for the analysis of the uncertainties corresponding to a complete measurement set up
(i.e. "a complete system") is also proposed in clause D.5. This approach addresses, in particular, the concept of
"sub-systems" and how to combine the uncertainties relating to each "sub-system". Such an approach could help in
cases where different units are to be used (e.g. dBs in one sub-system, linear terms in another).
A set of files (spread sheets) has been included in the present document, in order to support some of the examples given
and to help the user in the implementation of his own methodology.
5 Analysis of measurement uncertainty

This clause develops the approach to measurement uncertainty beyond the introduction given in clause 4. It details the
improvements to the analysis which the present document is proposing and presents solutions for all the identified
problems associated with the BIPM method for calculating measurement uncertainty in radiated measurements.
Clause 6 presents numerous worked examples which illustrate the application of the proposed new techniques.
In the beginning of this clause, a review is given of the BIPM method, along with an outline of where it is inadequate
for radiated measurements. The means of evaluation of type A and type B uncertainties are also given.
This is followed by a discussion of the units in which the uncertainties are derived and the technique for converting
standard deviations from logarithmic to linear quantities (% voltage or % power and vice versa) is presented. The
conversion technique allows all the individual uncertainty components in a particular test to be combined in the same
units and overcomes a major current day problem of asymmetric uncertainty limits (e.g. x + 2, -3 dB, as found in
edition 2 of ETR 028 [6]).
The clause concludes with clauses on deriving the expanded uncertainties in the case of Normal distributions, how
influence quantities are dealt with, calculating the standard deviation of random effects and an overall clause summary.
Theoretical and mathematical support for this clause can be found in annex D of TR 100 028-2 [8].
ETSI
26 ETSI TR 100 028-1 V1.4.1 (2001-12)
5.1 The BIPM method

Basic to the BIPM method is the representation of each individual uncertainty component that contributes to the overall
measurement uncertainty by an estimated standard deviation, termed standard uncertainty [1], with suggested symbol
u.
All individual uncertainties are categorized as either type A or type B.
Type A uncertainties, symbol ui, are estimated by statistical methods applied to repeated measurements, whilst type B
uncertainties, symbol uj, are estimated by means of available information and experience.
The combined standard uncertainty [1], symbol uc, of a measurement is calculated by combining the standard
uncertainties for each of the individual contributions identified. In the case where the underlying physical effects are
additive, this is done by applying the Root of the Sum of the Squares (the RSS) method under the assumption that all
contributions are stochastic i.e. independent of each other.
The table included in clause D.3.12 of TR 100 028-2 [8] provides the way in which should be handled contributions to
the uncertainty which correspond to physical effects which are not additive. Clause D.5 of the same annex provides an
overview of several more general methods.
The resulting combined standard uncertainty can then be multiplied by a constant kxx to give uncertainty limits
(bounds), termed expanded uncertainty [1]. When the combined standard uncertainty corresponds to a Normal
distribution (see clause 4.1.3) the expanded uncertainty corresponds to a confidence level of xx %.
This is the broad outline of the analysis technique employed in the present document, but there are numerous practical
problems when applying the basic BIPM rules to measurements, such as:
- how uncertainty contributions in different units (dB, % voltage, % power) can be combined;
- whether individual uncertainties are functions of the true value (e.g. Bit error ratios);
- how to deal with asymmetrically distributed individual uncertainties;
- how to evaluate confidence limits for those standard uncertainties which are not Normal by nature (see also
clause D.5.6.2 in TR 100 028-2 [8]).
These problem areas are discussed below and have resulted in modifications and extensions to the BIPM method. For
most cases, examples are given in clause 6.
In order to help understanding some of these questions and to bring some more theoretical support, annexes D and E
(found in TR 100 028-2 [8]) have been added to the third edition of the present document. Clause D.3 supports various
combinations (e.g. additive, multiplicative, etc.), conversions (e.g. to and from dBs) and functions (see clauses D.3.9
and D.3.11). A complete approach, encompassing the "BIPM method" is included in clause D.5 of TR 100 028-2 [8].
5.1.1 Type A uncertainties and their evaluation

Type A uncertainties are evaluated by statistical methods, estimating their standard deviations (corresponding to
"standard uncertainties"). These normally play a minor part in the combined standard uncertainty.
Annex D (in TR 100 028-2 [8]) shows that, in most cases, it is only the standard uncertainty that needs to be known in
order to find the combined uncertainty. In the BIPM approach, the shape of the individual distributions is relatively
unimportant. However, annex D shows how to combine the various individual distributions, when needed, and that the
result of a combination does not necessarily correspond to a Normal distribution. In such a case, the actual shape of the
resulting distribution may be fully relevant (see, in particular, clauses D.5.6.2.3 and D.5.6.2.4).
5.1.2 Type B uncertainties and their evaluation

Type B uncertainties are estimated by various methods.
Figure 5 illustrates a selection of uncertainty distributions which can often be identified in RF measurements.
ETSI
27 ETSI TR 100 028-1 V1.4.1 (2001-12)
-a 0 +a -a 0 +a
'U' Distribution Rectangular Distribution Normal (Gaussian) Distribution
Figure 5: Types of uncertainty distribution
Mismatch uncertainties have the "U" distribution, see annex G. The value of the uncertainty contribution is more likely
to be near the limits than to be small or zero. If the limits are ±a, the standard uncertainty is:
a
(see annex D)
2
Systematic uncertainties (e.g. those associated with the loss in a cable) are, unless the actual distribution is known,
assumed to have a rectangular distribution. The result of this assumption is that the uncertainty can take any value
between the limits with equal probability. If the limits are ±a, the standard uncertainty is:
a
(see annex D)
3
If the distribution used to model the uncertainty is a Normal distribution, it is characterized by its standard deviation
(standard uncertainty) (see annex D).
In the present document the standard uncertainties are symbolized by uj xx or uj description.
In all cases where the distribution of the uncertainty is unknown, the rectangular distribution should be taken as the
default model.
It will be noted that all the distributions illustrated in figure 5 are symmetrical about zero (clause D.1 addresses also
distributions showing an offset and/or which are not symmetrical). An unexpected complication in combining standard
uncertainty contributions may result from the use of different units, since a symmetrical standard uncertainty in %
voltage is asymmetrical in dB (and vice versa). Similarly for % power. This "major" complication (for any particular
test, the contributions may be in a variety of units) is the subject of clause 5.2. See also clause D.3 and in particular
clause D.3.10.7 of TR 100 028-2 [8].
5.2 Combining individual standard uncertainties in different

units
The BIPM method for calculating the combined standard uncertainty of any test involves combining the individual
standard uncertainties by the RSS method. If there are n individual standard uncertainty contributions to be combined,
the combined standard uncertainty is:
uc = u 2j1 + u 2j 2 + u 2j 3 + .... + u 2j ( n −1) + u 2jn + .... + ui21 + ui22 + ui23 + .... + ui2( n −1) + uin
2
(5.1)
However, this is correct only if all the individual contributions, represented by their standard uncertainties:
(1) combine by addition; and
(2) are expressed in the same units.
It does not matter whether the contributions are expressed in percent or logarithmic terms or any other terms as long as
these two conditions are fulfilled… noting that the result of the corresponding combination will be expressed in the
same way (see also conversions in clause D.3 and the discussion on the concept of sub-systems in clause D.5 of
TR 100 028-2 [8]).
ETSI
28 ETSI TR 100 028-1 V1.4.1 (2001-12)
To use formula 5.1 for standard uncertainties of individual contributions which combine by addition, linear terms only
i.e. voltage, percentage, etc., should be used. This is essential for the RSS combination to be valid. This is the case in
many measuring instruments.
To use formula 5.1 for standard uncertainties of individual contributions which combine by multiplication, logarithmic
terms only i.e. dB should be used as they can then be combined by addition. This is essential for the RSS combination
to be valid where uncertainty multiplication occurs. This is the case where gains and/or losses (i.e. attenuators,
amplifiers, antennas, etc.) are involved as well as under mismatch conditions where modules (i.e. attenuators, cables,
RF measuring instruments, etc.) are interconnected in RF measurements.
If all parameters and their associated standard uncertainties in a measurement are in the same unit and combine by
addition, the RSS method can be applied directly. The table in clause D.3.12 of TR 100 028-2 [8]) shows how to handle
other cases. Clause D.5 (in TR 100 028-2 [8]) discusses general methods usable in most cases.
For small (< 30 % or 2,5 dB) standard uncertainties however, both additive and multiplicative contributions can be
incorporated into the same calculation (with negligible error) provided they are converted to the same units prior to
calculating the combined standard uncertainty. The conversion factors are given in table 1. This is supported by the
theoretical analysis provided in annex D, clause D.3 and annex E (TR 100 028-2 [8]).
Annex E gives the justification for this statement by firstly mathematically converting the distribution of an individual
uncertainty from logarithmic to linear (and vice versa) and secondly comparing the standard deviation of the two
distributions before and after the conversion. One of the outcomes of annex E is that the conversion between linear and
logarithmic standard uncertainties can, under some conditions, be approximated by the first order mathematical
functions given in table 1.
As can be seen from annex E there are, however, some problems involved in converting distributions.
- It is not a linear procedure; the conversion factor is not only dependent on the magnitude of the standard
uncertainty, but it is also dependent on the shape of the distribution.
- The mean value of the converted uncertainty distribution is not necessarily zero, even if that was the case before
the conversion. However if the standard uncertainties to be converted are less than 2,5 dB, 30 % (voltage), or
50 % (power) the errors arising may be considered as negligible.
Table 1 shows the multiplicative factors to be used when converting standard uncertainties with a first order
approximation. As an example, if the standard uncertainty is 1,5 dB then this, converted to voltage %, gives a
corresponding standard uncertainty of 1,5 × 11,5 % = 17,3 %.
Table 1: Standard uncertainty conversion factors
Converting from standard Conversion factor To standard

uncertainties in …: multiply by: uncertainties in …:
dB 11,5 voltage %
dB 23,0 power %
power % 0,0435 dB
power % 0,5 voltage %
voltage % 2,0 power %
voltage % 0,0870 dB
It should be noted after any conversions that may be necessary before using equation 5.1, that the combined standard
uncertainty, uc, that results from the application of equation 5.1, does not, by itself give the expanded uncertainty limits
for a measurement.
When uc corresponds to a Normal distribution, these can be calculated (see clause 5.3) from uc (assumed in this case to
be in units of dB) as the 95 % confidence limits in dB of ±1,96 × uc (which is very asymmetric in linear terms).
ETSI
29 ETSI TR 100 028-1 V1.4.1 (2001-12)
Similarly, in voltage as ±1,96 × uc × 11,5 % (which is very asymmetric in dB terms).The major factor determining
whether the combined standard uncertainty, uc, will have the symmetrical dB interval or the symmetrical % interval (or
somewhere in-between) is whether the individual uncertainties combine by multiplication or by addition. In radiated
measurements as well as most conducted measurements where the RF level is of importance, the overwhelming
majority of the uncertainties combine by multiplication. It is, therefore, safe to assume that, in general, the resulting
uncertainty limits are symmetrical in logarithmic terms (dB). This assumption has been confirmed by computer
simulations on a large number of measurement models. This is also clear from the relations found in annex D.
5.3 Calculation of the expanded uncertainty values and

Student's t-distribution
This clause discusses two different problems, both relating to the handling of uncertainties, which have to be very
clearly identified and handled separately. Unfortunately, in the previous editions of the present document, this clause
had not been subdivided into two clauses.
The two clauses address:
- the situation where the statistical properties of a number of samples are to be evaluated; in this case, the
Student's t-distribution is a powerful tool allowing the evaluation of the performance of those properties; it can
be helpful in supporting the evaluation of properties of "type A uncertainties";
- the situation where only one measurement is performed, in conditions where the various sources of uncertainty
have been evaluated; as a result, the combined standard uncertainty of that measurement may be evaluated (see
clause D.5 of TR 100 028-2 [8]), and the knowledge of the shape of the distribution corresponding to that
combined uncertainty allows for changes in the confidence level.
5.3.1 Student's t-distribution

The Student's t-distribution gives coverage factors (i.e. multipliers) for measurements, whereby the confidence level of
a series of measurements can be calculated from a limited number of samples, assuming those samples have been taken
from a Normal distribution. The fewer the number of samples, the bigger the coverage factor for a given confidence
level.
For example:
- if a type A standard deviation is calculated on only 3 samples and the required confidence level is 95 % the
appropriate Student's t-factor is 3,18;
- if the standard deviation had been based on 20 samples, the factor would have been 2,09;
- for an infinite number of samples the multiplier would have been 1,96.
When using such an approach, any measurement should be repeated a large number of times.
In radio measurements, however, by using the approach recommended in the present document, only one measurement
is usually performed. As a result, the Student's t-distribution is of no help.
The Student's t-distribution can, however, be very useful for the statistical evaluation of the properties of individual
uncertainty components (i.e. type A uncertainties which may happen to be part of some test set up).
5.3.2 Expanded uncertainties

When the combined standard uncertainty, uc, has been calculated from equation 5.1 (or by any other method) and it can
be expected that the corresponding distribution is Normal, then, the uncertainty limits relate to a confidence level of
68,3 % (due to the properties of the Gaussian curve).
By multiplying uc by "a coverage factor" (or "an expansion factor") other confidence levels may be obtained when the
distribution corresponding to the combined standard uncertainty, uc, is Normal. Why?
ETSI
30 ETSI TR 100 028-1 V1.4.1 (2001-12)
When:
- all the individual sources of uncertainty are identified for all the tests;
- the distributions of the uncertainties of the individual sources are all known (or assumed);
- the maximum, worst-case values of all of the individual uncertainties are known.
Then, under these conditions, annex D of TR 100 028-2 [8] applies and the combined standard uncertainty can be
calculated (see clause D.5).
Assuming that the combined standard uncertainty corresponds to a Normal distribution then the magic factor of 1,96
applies: this is due to the shape of the Gaussian curve used to describe the distribution corresponding to the combined
uncertainty (see the interpretation in clause D.5.6.2 of TR 100 028-2 [8]).
As already indicated above (see clause 4.1.3), for a Gaussian shaped curve:
- a surface of 68 % (2 x 34 % , the value which can be found in some tables) corresponds to one standard
deviation (i.e. a combined standard deviation);
- a surface of 95 % (2 x 47,5 % , the value which can be found in some tables) corresponds to two standard
deviations (more precisely 1,96 standard deviations).
and the surface referred to above can be interpreted as the probability of the true value being within the stated
uncertainty bounds.
The probability of remaining inside this surface is, by definition, the confidence level.
It has to be made clear that, when the combination of the various components of the uncertainty corresponds to a
distribution which is not Normal, then other expansion factors apply in order to convert from one confidence level to
another. The values of these factors depend on the mathematical properties (i.e. the shape) of the corresponding
distribution.
It has to be made clear also that, as indicated in particular in annex D, when the number of components added (or
combined linearly) in order to obtain the uncertainty can be considered as an infinity, and under some other conditions,
then the distribution can be considered as Normal (based on the “Central Limit Theorem”). Under such conditions, the
factor 1,96 is valid (for a 95% confidence level). This is why it has been used extensively in the examples given in the
present document.
The usage of a value of 2 for this expansion factor has also been suggested (this would provide a confidence level of
95,45 % in the case of Normal distributions).
The tools given in annex D could allow for the calculation of the actual distribution corresponding to the combination of
various components for the uncertainty. Under such conditions, the appropriate expansion factors could also be
calculated, in the case where the distribution found would not have happened to be Normal.
5.4 Combining standard uncertainties of different parameters,

where their influence on each other is dependant on the
EUT (influence quantities)
In many measurements, variations in the influence quantities, intermediate test results or test signals can affect the
uncertainty of the measurand in ways that may be functions of the characteristics of the EUT and other instrumentation.
ETSI
31 ETSI TR 100 028-1 V1.4.1 (2001-12)
It is not always possible to fully characterize test conditions, signals and measurands. Uncertainties are related to each
of them. These uncertainties may be well known, but their influence on the combined standard uncertainty depends on
the EUT. Uncertainties related to general test conditions are:
- ambient temperature;
- the effect of cooling and heating;
- power supply voltage;
- power supply impedance;
- impedance of test equipment connectors (VSWR).
Uncertainties related to applied test signals and measured values are:
- level;
- frequency;
- modulation;
- distortion;
- noise.
The effect of such uncertainties on the test results can vary from one EUT to another. Examples of the characteristics
that can affect the calculation of the uncertainties are:
- receiver noise dependency of RF input signal levels;
- impedance of input and output connectors (VSWR);
- receiver noise distribution;
- performance dependency of changes of test conditions and test signals;
- modulator limiting function e.g. maximum deviation limiting;
- system random noise.
If the appropriate value for each characteristic has not been determined for a particular case, then the values listed in
TR 100 028-2 [8] table F.1 should be used. These values are based on measurements made with several pieces of
equipment and are stated as mean values associated with a standard uncertainty reflecting the spread from one EUT to
another.
When the EUT dependent uncertainties add to the combined standard uncertainty, the RSS method of combining the
standard uncertainties is used, but in many calculations the EUT dependency is a function that converts uncertainty
from one part of the measurement configuration to another. In most cases the EUT dependency function can be assumed
to be linear; therefore the conversion is carried out by multiplication, as shown in the theoretical analysis provided in
clause D.4 of TR 100 028-2 [8].
The standard uncertainty to be converted is uj 1. The mean value of the influence quantity is A and its standard
uncertainty is uj a. The resulting standard uncertainty uj converted of the conversion is:
u jconverted = u 2j1( A2 + u 2j a ) (5.2)
The standard uncertainty of this contribution is then looked upon as any other individual component and is combined
accordingly (see annex D). A fully worked example of an influence quantity is given in clause 6.4.6. The conditions
under which the expression 5.2 is valid can also be found in clause D.4 of TR 100 028-2 [8].
ETSI
32 ETSI TR 100 028-1 V1.4.1 (2001-12)
If the function is not linear another solution must be found:
- the theoretical relation between the influence quantity and its effect has to be determined;
- the expressions providing the conversion can then be found based on the table contained in clause D.3.12 of
TR 100 028-2 [8].
When the theoretical relation between the influence quantity and its effect is not known, the usage of a simple
mathematical model can be tried. In this case, an attempt can be made in order to determine the numerical values of the
parameters of the model by some statistical method (see also clause D.5.4 in TR 100 028-2 [8]).
In all cases, it is recommended to determine first the mathematical relation between the parameters, and only after try
and find the appropriate numerical values. As a consequence, tables similar to table F.1 in annex F (TR 100 028-2 [8])
should also include the mathematical relation between the parameters for each entry (for further details, see clause
D.4.2.1.2 in TR 100 028-2 [8]).
5.5 Uncertainties and randomness

The major difficulty behind this clause is to understand exactly what “randomness uncertainty” is supposed to cover in
this context (i.e. what this clause or contribution is expected to cover): the BIPM method and the corresponding analysis
is supposed to cover all components of the uncertainty, so it is fundamental to understand what is left over for the
"uncertainty of randomness", in order to avoid taking into account the same effects twice, under different names (in a
complex set up)…
The standard uncertainty of randomness can be evaluated by repeating a measurement (e.g. of a particular component of
the measurement uncertainty).
The first step is to calculate the arithmetic mean or average of the results obtained.
The spread in the measured results reflects the merit of the measurement process and depends on the apparatus used, the
method, the sample and sometimes the person making the measurement. A more useful statistic, however, is the
standard uncertainty σi of the sample. This is the root mean square of the differences between the measured values and
the arithmetic mean of the samples.
If there are n results for xm where m = 1, 2, ..., n and the sample mean isx, then the standard deviation σi is:
n 2
 _
∑
1
σi = x
 m
− x

(5.3)
n
m =1  
This should not be confused with the standard deviation of the A uncertainty being investigated. It only covers n
samples.
If further measurements are made, then for each sample of results considered, different values for the arithmetic mean
and standard deviation will be obtained. For large values of n these mean values approach a central limit value of a
distribution of all possible values. This distribution can usually be assumed, for practical purposes, to be a Normal
distribution.
From the results of a relatively small number of measurements an estimate can be made of the standard deviation of the
whole population of possible values, of which the measured values are a sample.
Estimate of the standard deviation σi/:
n 2
 _
∑
1
σ i/ = ( ) x
 m
− x

(5.4)
n-1
m=1  
ETSI
33 ETSI TR 100 028-1 V1.4.1 (2001-12)
A practical form of this formula is:
X2
Y-
σ i/ = n (5.5)
n -1
where X is the sum of the measured values and Y is the sum of the squares of the measured values.
It will be noted that the only difference between σi/ and σi is in the factor 1/ (n-1) in place of 1/n, so that the difference
becomes smaller as the number of measurements is increased. A similar way of calculating the standard deviation of a
discrete distribution can be derived from this formula.
In this case X is the sum of the individual values from the distribution times their probability, and Y is the sum of the
square of the individual values times their probability.
If the distribution has m values xi, each having the probability p (xi):
m
X = ∑ x p(x )
i =1
i i (5.6)
and
m
Y= ∑
i =1
xi2 p(xi ) (5.7)
The standard uncertainty is then:
σi = Y − X 2 (5.8)
When measured results are obtained as the arithmetic mean of a series of n (independent) measurements the standard
uncertainty is reduced by a factor √n thus:
σ 1/
σi = (5.9)
n
This is an efficient method of reducing measurement uncertainty when making noisy or fluctuating measurements, and
it applies both for random uncertainties in the measurement configuration and the EUT. Having established the standard
deviation, this is directly equated to the standard uncertainty:
ui = σi
As the uncertainty due to random uncertainty is highly dependent on the measurement configuration and the
test method used it is not possible to estimate a general value.
Each laboratory must by means of repetitive measurements estimate their own standard uncertainties characterizing the
randomness involved in each measurement. Once having done this, the estimations may be used in future measurements
and calculations.
NOTE: See also the note found in clause 6.4.7 concerning the usage of this component.
5.6 Summary of the recommended approach

The basic BIPM method, with specific modifications, remains the adopted approach used for the calculation of
combined standard and expanded uncertainty in the examples given in this report. That is to say that once all the
individual standard uncertainties in a particular measurement have been identified and given values, they are combined
by the RSS method provided they combine by addition and are in the same units (otherwise, methods such as
those detailed in annex D, e.g. in clause D.5, have to be used).
ETSI
34 ETSI TR 100 028-1 V1.4.1 (2001-12)
In order to ensure that this proviso is satisfied as often as possible, the present document supplies the factors necessary
to convert standard uncertainties in linear units to standard uncertainties in logarithmic units (and vice versa). The
present document also shows that small additive standard uncertainties (% V, % power) can be combined with
multiplicative standard uncertainties (dB) in the RSS manner with, hopefully, negligible error.
Having derived the combined standard uncertainty, an expanded uncertainty for 95 % confidence levels can then be
derived, when the corresponding distribution is Normal, by multiplying the result by the expansion factor of 1,96. The
multiplication by this factor (or simply by a factor equal to 2) is to be done in all cases, in order to obtain the expanded
uncertainty. However, if the corresponding distribution is not Normal, then the resulting confidence level is not
necessarily 95 % (see clause D.5.6.2 of TR 100 028-2 [8]). In all cases, however, the actual confidence level can be
calculated, once the distribution corresponding to the combination of all uncertainty components has been calculated.
Clause D.3, in TR 100 028-2 [8], provides the equations allowing for the calculation of this combined distribution.
The practical implementation of this modified BIPM approach, adopted throughout the present document, is for each
test method (including the verification procedures) to have appended to it a complete list of the individual uncertainty
sources that contribute to each stage of the test. Magnitudes of the standard uncertainties can then be assigned to these
individual contributions by consulting annex A (converting from linear units to dB, if necessary). All uncertainties are
in dB units since the great majority of the individual contributions in radiated measurements are multiplicative i.e. they
add in dB terms.
In those cases in which annex A instructs that the values of the uncertainty contributions be taken from a manufacturer's
data sheet, that data should be taken over as broad a frequency band as possible. This type of approach avoids the
necessity of calculating the combined standard uncertainty every time the same test is performed for different
EUT.
6 Examples of uncertainty calculations specific to radio

equipment
6.1 Mismatch
In the following the Greek letter Γ means the complex reflection coefficient. ρx is the magnitude of the reflection
coefficient: ρx =  Γx.
Where two parts or elements in a measurement configuration are connected, if the matching is not ideal, there will be an
uncertainty in the level of the RF signal passing through the connection. The magnitude of the uncertainty depends on
the VSWR at the junction of the two connectors.
The uncertainty limits of the mismatch at the junction are calculated by means of the following formula:
Mismatch limits = |Γgenerator| × |Γload| × |S21| × |S12| × 100 % Voltage (6.1)
where:
- |Γgenerator| is the modulus of the complex reflection coefficient of the signal generator;
- |Γload| is the modulus of the complex reflection coefficient of the load (receiving device);
- | S21| is the forward gain in the network between the two reflection coefficients of interest;
- | S12| is the backward gain in the network between the two reflection coefficients of interest.
NOTE: S21 and S12 are set to 1 if the two parts are connected directly. In linear networks S21 and S12 are identical.
The distribution of the mismatch uncertainty is U-shaped, If the uncertainty limits are ± a, the standard uncertainty is:
Γgenerator × Γload × S 21 × S12 × 100 %

u j mismatch:individual = Voltage % (6.2)
2
ETSI
35 ETSI TR 100 028-1 V1.4.1 (2001-12)
This can be converted into equivalent dB by dividing by 11,5 (see clause 5.2):
Γgenerator × Γload × S 21 × S12 × 100 %

u j mismatch:individual = dB (6.3)
2 × 11,5
If there are several connections in a test set-up, they will all interact and contribute to the combined mismatch
uncertainty. The method of calculating the combined mismatch uncertainty is fully explained in annex G.
In conducted measurements, when calculating the mismatch uncertainty at the antenna connector of the EUT, the
reflection coefficient of the EUT is required. In this case, the laboratory should either measure it in advance or use the
reflection coefficients given in TR 100 028-2 [8], table F.1.
6.2 Attenuation measurement

In many measurements the absolute level of the RF signal is part of the measured result. The RF signal path attenuation
must be known in order to apply a systematic correction to the result. The RF signal path can be characterized using the
manufacturers' information about the components involved, but this method can result in unacceptably large
uncertainties.
Another method is to measure the attenuation directly by using, for example, a signal generator and a receiving device.
To measure the attenuation, connect the signal generator to the receiving device and read the reference level (A), see
figure 6, and then insert the unknown attenuation, repeat the measurement and read the new level (B), see figure 7.
Γ generator Γ load Receiving

Generator
device
Figure 6: Measurement of level (A)
In figure 6, Γgenerator is the complex reflection coefficient of the signal generator and Γload is the complex reflection
coefficient of the load (receiving device).
Γgenerator Γinput Γoutput Γload Receiving

Generator Attenuator
device
Figure 7: Measurement of level (B)
In figure 7, Γgenerator is the complex reflection coefficient of the signal generator, Γload is the complex reflection
coefficient of the load (receiving device), Γinput is the complex reflection coefficient of the attenuator input, Γoutput is
the complex reflection coefficient of the attenuator output.
The attenuation is calculated as A/B if the readings are linear values or A-B if the readings are in dB.
Using this method, four uncertainty sources need to be considered. Two sources concern the receiving device, namely
its absolute level (if the input attenuation range has been changed) and its linearity. The other two sources are the
stability of the signal generator output level (which contributes to both stages of the measurement) and mismatch caused
by reflections at both the terminals of the network under test and the instruments used. The absolute level, linearity and
stability uncertainties can be obtained from the manufacturers data sheets, but the mismatch uncertainty must be
estimated by calculation.
For this example, we assume that an attenuator of nominally 20 dB is measured at a frequency of 500 MHz by means of
a signal generator and a receiving device. The magnitude of the reflection coefficient of the generator |Γgenerator| is 0,2,
the magnitude of the reflection coefficient of the receiving device |Γload| is 0,3 and the magnitude of the reflection
coefficients of the attenuator |Γinput| and |Γoutput| are 0,05.
Since the mismatch uncertainty of the attenuation measurement is different in figure 7 to that in figure 6, it therefore has
to be calculated (for figure 6 and figure 7) and both values included in the combined mismatch uncertainty as shown
below.
ETSI
36 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mismatch uncertainty:
Reference measurement: The signal generator is adjusted to 0 dBm and the reference level A is measured on the
receiving device. Using equation 6.1 with S21 = S12 = 1, and taking the standard uncertainty,
uj mismatch: reference measurement:
0,2 × 0,3 × 100 4 ,24

u j mismatch:referencemeasurement = % = 4,24 % ≈ = 0,37 dB
2 11,5
Attenuator measurement: The attenuator is inserted and a level (B) = -20,2 dB is measured after an input attenuation
range change on the receiving device.
NOTE: The measured attenuation is 20,2 dB, for which S21 = S12 = 0,098.
The following three components comprise the uncertainty in this part of the measurement:
- the standard uncertainty of the mismatch between the signal generator and the attenuator:
0,2 × 0,05 × 100

u j mismatch:generatortoattenuator = % = 0,71 %
2
- the standard uncertainty of the mismatch between the attenuator and the receiving device:
0 ,3 × 0 ,05 × 100
u j mismatch:attenuatortoreceivingdevice = % = 1,06 %
2
- the standard uncertainty of the mismatch between the signal generator and the receiving device:
0,3 × 0,2 × 0,098 2 × 100

u j mismatch:generatortoreceivingdevice = % = 0,041 %
2
The combined standard uncertainty of the mismatch of the attenuation measurement uc mismatch: att. measurement, is
calculated by RSS (see clause 5.2) of the individual contributions.
1,28
u c mismatch:att.measurement = 0,712 + 1,06 2 + 0,0412 = 1,28 % ≈ = 0,11 dB
11,5
A comparison of uj mismatch: reference measurement (0,37 dB) and uc mismatch: att. measurement (0,11 dB) shows clearly the
impact of inserting an attenuator between two mismatches.
Other components of uncertainty:
Reference measurement: The stability of the signal generator provides the only other uncertainty in the present
document. The receiving device contributes no uncertainty here since only a reference level is being set for comparison
in the attenuation measurement stage.
The output level stability of the signal generator is taken from the manufacturer's data sheet as 0,10 dB which is
assumed (since no information is given) to be rectangularly distributed (see clause 5.1). Therefore the standard
uncertainty, uj signal generator stability, is:
0,10
u j signal generatorstability = = 0,06 dB
3
Therefore, the combined standard uncertainty, uc reference measurement, for the reference measurement is:
u c referencemeasurement = u 2j mismatch:referencemeasurement + u 2j signal generatorstability = 0,37 2 + 0 ,06 2 = 0,37 dB
Attenuation measurement: Here the output stability of the signal generator as well as absolute level uncertainty of the
receiving device (the input attenuation range has changed) contribute to the uncertainty. However as a range change has
occurred there is no linearity contribution as this is included in the absolute level uncertainty of the receiver.
ETSI
37 ETSI TR 100 028-1 V1.4.1 (2001-12)
The signal generator stability, uj signal generator stability, has the same value as for the reference measurement, whilst the
uncertainty for the receiving device is given in the manufacturer's data sheet as 1,0 dB absolute level accuracy. A
rectangular distribution is assumed for the absolute level accuracy so the standard uncertainty, uj signal generator level, of
its uncertainty contribution is:
1,00
u j signal generatorlevel = = 0,58 dB
3
The uncertainty contribution of the linearity of the receiving device uj linearity is zero.
Therefore the combined standard uncertainty, uc att. measurement, for the attenuation measurement is:
uc att.measurement = uc2 mismatch:att.measurement + u 2j signal generator stability + u 2j signal generatorlevel + u 2j linearity
= 0 ,08 2 + 0,06 2 + 0 ,58 2 + 0,00 2 = 0,59 dB
So, for the complete measurement, the combined standard uncertainty, uc measurement, is given by:
u cmeasurement = u c referencemeasurement 2 + u c att.measurement 2 = 0 ,37 2 + 0,59 2 = 0,70 dB
The expanded uncertainty is ±1,96 × 0,70 = ±1,37 dB at a 95 % confidence level.
This is an exaggerated example. Smaller uncertainty is possible if a better receiving device is used.
6.3 Calculation involving a dependency function

The specific dependency function is the relationship between the RF signal level at the EUT antenna connector (dB) to
the uncertainty of the measurement of SINAD at the EUT's audio output i.e. how does SINAD measurement uncertainty
relate to RF level uncertainty at the EUT antenna connector.
The following example is based on a typical ETR 028 [6] type (conducted) RF measurement for clarity. The sensitivity
of a receiving EUT is measured. The outline of the measurement is as follows. The RF level at the input of the receiver
is continuously reduced until a SINAD measurement of 20 dB is obtained, see figure 8.
The result of the measurement is the RF signal level causing 20 dB SINAD at the audio output of the receiver.
|ρgenerator |= 0,30 |S11 | = 0,07

EUT
Signal cable SINAD
generator ± 0,5 dB EUT dependancy function meter
mean value = 1
± 1,0 dB |S 22| = 0,07 |ρEUT|= 0,4 standard deviation = 0,3 ± 1,0 dB
Figure 8: Typical measurement configuration
The combined standard uncertainty is calculated as follows.
For the mismatch uncertainty (annex G):
Generator: Output reflection coefficient: |ρgenerator| = 0,30
Cable: Input and output reflection coefficients: |S11| and |S22| = 0,07
Attenuation: 1 dB = |S21| = |S12| = 0,891
EUT: Input reflection coefficient: |ρEUT| = 0,4
ETSI
38 ETSI TR 100 028-1 V1.4.1 (2001-12)
All these contributions are U distributed. There are three contributions:
- the standard uncertainty of the mismatch between the signal generator and the cable:
0,30×0,07×100%
u j mismatch:signal generatortocable = = 0,13 dB
2 ×11,5
- the standard uncertainty of the mismatch between the cable and the EUT:
0,4×0,07×100%
u j mismatch:cableto EUT = = 0,17 dB
2 ×11,5
- the standard uncertainty of the mismatch between the signal generator and the EUT:
0,3×0,4×0,8912 × 100%
u j mismatch:signal generatorto EUT = = 0,59 dB
2 ×11,5
- the combined standard uncertainty of the mismatch:
u c mismatch = 0,132 + 0,17 2 + 0,59 2 = 0,63 dB
uc mismatch = 0,63 dB
The uncertainty due to the absolute output level of the signal generator is taken as ±1,0 dB (from manufacturers data).
As nothing is said about the distribution, a rectangular distribution in logs is assumed (see clause 5.1), and the standard
uncertainty is:
uj signal generator level = 0,58 dB
The uncertainty due to the output level stability of the signal generator is taken as ±0,02 dB (from manufacturer's data).
As nothing is said about the distribution, a rectangular distribution in logs is assumed (see clause 5.1), and the standard
uncertainty is:
uj signal generator stability = 0,01 dB
The uncertainty due to the insertion loss of the cable is taken as ±0,5 dB (from calibration data). As nothing is said
about the distribution, a rectangular distribution in logs is assumed, and the standard uncertainty is:
uj cable loss = 0,29 dB
Dependency function uncertainty calculation:
The uncertainty due to the SINAD measurement corresponds to an RF signal level uncertainty at the input of the
receiving EUT.
The SINAD uncertainty from the manufacturer's data is ±1 dB which is converted to a standard uncertainty of 0,577 dB.
The dependency function converting the SINAD uncertainty to RF level uncertainty is found from TR 100 028-2 [8],
table F.1. It is given as a conversion factor of 1,0 % (level)/ % (SINAD) with an associated standard uncertainty of 0,3.
The SINAD uncertainty is then converted to RF level uncertainty using formula 5.2:
( )
u j RF level(converted ) = 0 ,577 2 × 1,0 2 + 0,3 2 = 0,60 dB
The RF level uncertainty caused by the SINAD uncertainty and the RF level uncertainty at the input of the receiver is
then combined using the square root of the sum of the squares method to give the combined standard uncertainty.
uc measurement = uc2 mismatch + u 2j signal generatorlevel + u 2j signal generator stability + u 2j cableloss + u 2j RF level (converted )
= 0,63 2 + 0,58 2 + 0,012 + 0,29 2 + 0 ,60 2 = 1,08 dB
ETSI
39 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.4 Measurement of carrier power

The example test is a conducted measurement.
6.4.1 Measurement set-up

The EUT is connected to the power meter via a coaxial cable and two power attenuators, one of 10 dB and one of 20 dB
(see figure 9).
Transmitter Cable 10 dB power 20 dB power Power

Sensor
under test attenuator attenuator meter
Figure 9: Measurement set-up
The nominal carrier power is 25 W, as a result the power level at the input of the power sensor is (nominally) 25 mW.
The carrier frequency is 460 MHz and the transmitter is designed for continuous use.
6.4.2 Method of measurement

The transmitter is in an environmental chamber adjusted to +55°C. The attenuators and the power sensor are outside the
chamber.
Prior to the power measurement the total insertion loss of cable and attenuators is measured.
The attenuation measurements are done using a generator and a measuring receiver and two 6 dB attenuators with small
VSWR.
Also the power sensor is calibrated using the built in power reference.
The result of the measurement is the power found as the average value of 9 readings from the power meter, corrected
for the measured insertion loss.
6.4.3 Power meter and sensor module

The power meter uses a thermocouple power sensor module and contains a power reference.
Power reference level:
Power reference level uncertainty: ±1,2 % power.
As nothing is stated about the distribution it is assumed to be rectangular and the standard uncertainty is converted
from % power to dB by division with 23,0 (see clause 5.2).
1,2
Standard uncertainty u j referencelevel = = 0,030 dB
3 × 23,0
Mismatch whilst measuring the reference:
- Reference source VSWR: 1,05 (d): ρreference source = 0,024;
- Power sensor VSWR: 1,15 (d): ρload = 0,07.
Using formula 6.3 the standard uncertainty of the mismatch is:
0 ,024×0 ,07 × 100%

u j mismatch:reference = = 0,010 dB
2 × 11,5
ETSI
40 ETSI TR 100 028-1 V1.4.1 (2001-12)
Calibration factors:
Calibration factor uncertainty = ±2,3 % power.
As nothing is stated about the distribution it is assumed to be rectangular. The standard uncertainty is converted from %
power to dB by division with 23,0.
2 ,3
standard uncertainty u j calibration factor = = 0,058 dB
3×23,0
Range to range change:
Range to range uncertainty (one change) = ±0,5 % power.
As nothing is stated about the distribution it is assumed to be rectangular. The standard uncertainty is converted from %
power to dB by division with 23,0.
0 ,25
standard uncertainty u j rangechange = = 0,006 dB
3×23,0
Noise and drift is negligible at this power level and can be ignored.
Combined standard uncertainty of the power meter and sensor:
Using formula 5.1:
uc meter and sensor = u 2j referencelevel + u 2j mismatch: reference + u 2j calibration factor + u 2j rangechange
uc meter and sensor = 0,032 + 0,0102 + 0,0582 + 0,0062 = 0,066 dB
6.4.4 Attenuator and cabling network

Standing wave ratios involved in the attenuation measurement (taken from manufacturers data):
- Signal generator: VSWR ≤ 1,5 ρ = 0,200
- Power sensor: VSWR ≤ 1,15 ρ = 0,070
- 6 dB attenuators: VSWR ≤ 1,2 ρ = 0,091
- 10 dB power attenuator: VSWR ≤ 1,3 ρ = 0,130
- 20 dB attenuator: VSWR ≤ 1,25 ρ = 0,111
- Cable: VSWR ≤ 1,2 ρ = 0,091

Nominal attenuations converted to linear values:
- 6 dB = S21=S12= 0,500;
- 10 dB = S21=S12= 0,316;
- 20 dB = S21=S12= 0,100;
- 0,3 dB = S21=S12= 0,966 (assumed cable attenuation in the uncertainty calculations).
The attenuation measurement is carried out using a signal generator and a measuring receiver. In order to have a low
VSWR two 6 dB attenuators with low reflection coefficients are inserted.
The measurement of the attenuation in the attenuator and cabling network is carried out by making a reference
measurement (figure 10). The measurement receiver reading is "A" dBm.
ETSI
41 ETSI TR 100 028-1 V1.4.1 (2001-12)
Then the cables and the attenuators are inserted. First the cable and the 10 dB power attenuator is inserted between the
two 6 dB attenuators, and a new reading "B" dBm is recorded (see figure 11).
Finally the 20 dB attenuator is inserted between the two 6 dB attenuators, and the reading "C" dBm is recorded (see
figure 12).
The total attenuation is then ("A"-"B") dB + ("A"-"C") dB.
6.4.4.1 Reference measurement

Figure 10 details the components involved in this reference measurement.
RF signal 6 dB (1) 6 dB (2) Power meter

generator attenuator attenuator and sensor
Figure 10: The reference measurement
The individual mismatch uncertainties between the various components in figure 10 are calculated using formula 6.3:
- the standard uncertainty of the mismatch between the signal generator and 6 dB attenuator (1):
0,2 × 0,091× 100 %

u j mismatch:generatorto6dBatt. = = 0,112 dB
2 × 11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (1) and 6 dB attenuator (2):
0,091×0,091×100%
u j mismatch:6dBatt.1to6dBatt.2 = = 0,051 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (2) and power sensor:
0 ,091×0,07×100%
u j mismatch:6dBatt.to power sensor = = 0,039 dB
2 ×11,5
0,2×0,091×0,5 2 × 100%
u j mismatch:generatorto6dBatt.2 = = 0,028 dB
2 ×11,5
0,091×0,07×0,5 2 × 100%
u j mismatch:6dBatt.1to power sensor = = 0,010 dB
2 ×11,5
- the standard uncertainty of the mismatch between the signal generator and power sensor:
0 ,2×0 ,07×0,52 ×0 ,52 ×100%

u j mismatch:generatorto power sensor = = 0,005 dB
2 ×11,5
It can be seen that the mismatch uncertainty between the RF signal generator and the 6 dB attenuator (1)
uj generator to 6 dB att 1, and the mismatch uncertainty between the 6 dB attenuator (2) and the power sensor
uj 6 dB att. 2 to power sensor, add to both the reference measurement and the measurements with the unknown attenuators
inserted.
ETSI
42 ETSI TR 100 028-1 V1.4.1 (2001-12)
It is the result of the methodology adopted in annex D that these terms cancel and hence do not contribute to the
combined standard uncertainty of the final result. The reference measurement mismatch uncertainty uj mismatch: reference
(formula 5.1):
u j mismatch:reference = u 2j 6 dB att.1to 6dB att.2 + u 2j generator to6 dB att.2 + u 2j 6dB att.1to power sensor + u 2j generator to power sensor
u j mismatch:reference = 0 ,0512 + 0,0282 + 0,0102 + 0 ,0052 = 0,059 dB
NOTE: If the two uncertainties of the generator and the power sensor did not cancel due to the methodology, the
calculated reference measurement uncertainty would have been 0,131 dB.
6.4.4.2 The cable and the 10 dB power attenuator

Figure 11 shows the clause of the reference set-up which concerns this part of the calculation.
RF signal 6 dB (1) Cable 10 dB power 6 dB (2) Power meter

generator attenuator attenuator attenuator and sensor
Figure 11: The cable and the 10 dB power attenuator
The individual uncertainties are calculated using formula 6.3:
0,2 × 0 ,091×100%
u j mismatch:generatorto6dB att. = = 0,112 dB
2 × 11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (1) and cable:
0 ,091×0,091×100%
u j mismatch:6dBatt.1tocable = = 0,051 dB
2 ×11,5
- the standard uncertainty of the mismatch between the cable and 10 dB power attenuator:
0,091×0,130×100%
u j mismatch:cableto10dBatt. = = 0,073 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 10 dB attenuator and the 6 dB attenuator (2):
0 ,130×0,091×100%
u j mismatch:10dBatt.to6dBatt.2 = = 0,073 dB
2 ×11,5
0 ,091×0,07×100%
2 ×11,5
- the standard uncertainty of the mismatch between the signal generator and cable:
0 ,200×0,091×0,5 2 × 100%
u j mismatch:generatortocable = = 0,028 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (1) and 10 dB power attenuator:
0,091×0 ,130×0,966 2 ×100%

u j mismatch:6dBatt.1to10dB att. = = 0,068 dB
2 ×11,5
ETSI
43 ETSI TR 100 028-1 V1.4.1 (2001-12)
- the standard uncertainty of the mismatch between the cable and 6 dB attenuator (2):
0,091×0 ,091×0,316 2 ×100%

u j mismatch:cableto6dBatt.2 = = 0,005 dB
2×11,5
- the standard uncertainty of the mismatch between the 10 dB power attenuator and the power sensor:
0 ,130×0,070×0,500 2 ×100%
2 × 11,5
- the standard uncertainty of the mismatch between the signal generator and 10 dB power attenuator:
0,200×0 ,130×0,500 2 ×0,966 2 ×100%

2 ×11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (1) and 6 dB attenuator (2):
0 ,091×0 ,091×0 ,966 2 ×0,316 2 ×100%

u j mismatch:6dBatt.1to6dBatt.2 = = 0,005 dB
2 ×11,5
- the standard uncertainty of the mismatch between the cable and power sensor:
0 ,091×0,070×0,316 2 ×0,500 2 ×100%

u j mismatch:cableto power sensor = = 0,001 dB
2 ×11,5
0 ,200×0,091×0 ,500 2 ×0,966 2 ×0,316 2 ×100%

u j mismatch:generatorto6dBatt.2 = = 0,003 dB
2 ×11,5
0,091×0 ,070×0,966 2 ×0 ,316 2 ×0,500 2 ×100%

u j mismatch:6dBatt.1to power sensor = = 0,001 dB
2 ×11,5
- the standard uncertainty of the mismatch between the signal generator and power sensor:
0 ,200×0,070×0 ,500 2 ×0 ,966 2 ×0 ,316 2 ×0 ,500 2 ×100%

u j mismatch:generatorto power sensor = = 0,000 dB
2 ×11,5
The combined mismatch uncertainty when measuring the power level when the cable and the 10 dB power attenuator is
inserted is the RSS of all these components except uj mismatch: generator to 6 dB attenuator
and uj mismatch: 6 dB attenuator to power sensor:
uc mismatch:10dB and cable = u 2j mismatch:6 dB att.1to cable + ...... + u 2j mismatch:generatorto power sensor
.
u c mismatch:10dBand cable = 0,0512 + 0,0732 + .... + 0,0012 + 0,000 2 = 0,142 dB
The combined standard uncertainty of the mismatch when measuring the 10 dB attenuator and cable is:
uc mismatch:10dB and cablemeasurement = uc2mismatch:10 dB att and cable + uc2mismatch:reference

.
u c mismatch:10dBand cablemeasurement = 0,142 2 + 0,059 2 = 0,154 dB
ETSI
44 ETSI TR 100 028-1 V1.4.1 (2001-12)
The combined standard uncertainty of the mismatch uc mismatch: 10 dB and cable is 0,154 dB.
NOTE: The result would have been the same if only the 6 dominant terms were taken into account. This
illustrates that combinations of reflection coefficients separated by attenuations of 10 dB or more can
normally be neglected. The exceptions may be in cases where one or both of the reflection coefficients
involved are approaching 1,0 - which can be the case with filters or antennas outside their working
frequencies.
6.4.4.3 The 20 dB attenuator

Figure 12 shows the clause of the set-up which concerns this part of the calculation.
RF signal 6 dB (1) 20 dB power 6 dB (2) Power meter

generator attenuator attenuator attenuator and sensor
Figure 12: The 20 dB attenuator
In this part only terms separated by less than 10 dB are taken into account.
The individual uncertainties are calculated using formula 6.3:
0,2 × 0,091× 100%

2 × 11,5
- the standard uncertainty of the mismatch between the 6 dB attenuator (1) and 20 dB attenuator:
0,091×0,111×100%
u j mismatch:6dBatt.1to20dBatt. = = 0,062 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 20 dB attenuator and 6 dB attenuator (2):
0,111×0,091×100%
u j mismatch:20dBatt.to6dBatt.2 = = 0,062 dB
2 ×11,5
0 ,091×0,07×100%
u j mismatch:6 dB att.to power sensor = = 0,039 dB
2 ×11,5
- the standard uncertainty of the mismatch between the signal generator and 20 dB attenuator:
0,200×0 ,111×0,5002 ×100%

u j mismatch:generaorto 20dB att. = = 0,034 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 20 dB attenuator and power sensor:
0,111×0 ,070×0,5002 ×100%

u j mismatch:20dB att.to power sensor = = 0,012 dB
2 ×11,5
The rest of the combinations are not taken into account because the insertion losses between them are so high, that the
values are negligible:
- 6 dB attenuator (1) and 6 dB attenuator (2);
- signal generator and 6 dB attenuator (2);
ETSI
45 ETSI TR 100 028-1 V1.4.1 (2001-12)
- 6 dB attenuator (1) and measuring receiver;
- signal generator and measuring receiver.
The combined standard uncertainty of the mismatch when measuring the attenuation of the 20 dB attenuator is the RSS
of these 4 individual standard uncertainty values:
u j mismatch:20dB = 0,062 2 + 0,062 2 + 0,034 2 + 0,012 2 = 0,095dB
The combined standard uncertainty of the mismatch involved in the 20 dB attenuator measurement is:
uc mismatch:20dB measurement = uc2 mismatch:20dB + uc2 mismatch:reference
uc mismatch:20dB measurement = 0 ,0952 + 0 ,0592 = 0,112 dB
NOTE: If the two 6 dB attenuators had not been inserted, the result would have been 0,265 dB.
6.4.4.4 Instrumentation
Linearity of the measuring receiver is ±0,04 dB (from manufacturers data) as nothing is said about the distribution, a
rectangular distribution in logs is assumed and the standard uncertainty is calculated:
0,04
standard uncertainty u j receiverlinearity = = 0,023 dB
3
6.4.4.5 Power and temperature influences

Temperature influence: 0,0001 dB/degree (from manufacturers data), which is negligible, the power influence for the
10 dB attenuator is 0,0001 dB/dB × Watt (from manufacturers data) which gives 0,0001 × 25 × 10 = 0,025 dB as
nothing is said about the distribution, a rectangular distribution in logs is assumed and the standard uncertainty is
calculated:
0,025
u j powerin fluence10dB = = 0,014 dB
3
The power influence for the 20 dB attenuator is 0,001 dB/dB × Watt (from manufacturers data) which gives
0,001 × 2,5 × 20 = 0,05 dB as nothing is said about the distribution, a rectangular distribution in logs is assumed and the
standard uncertainty is calculated:
0,050
u j poweri nfluence20dB = = 0,028 dB
3
6.4.4.6 Collecting terms

10 dB attenuator and cabling network uncertainty:
uc 10 dB attenuatorand cable = uc2 mismatch + u 2j receiverlinearity + u 2j poweri nfluence10 dB
uc 10dB attenuatorand cable = 0,1542 + 0,0232 + 0,0142 = 0,156 dB
ETSI
46 ETSI TR 100 028-1 V1.4.1 (2001-12)
20 dB attenuator and cabling network uncertainty:
uc 20dB attenuator = uc2 mismatch + u 2j receiverlinearity + u 2j poweri nfluence20dB
uc 20dB attenuator = 0,1122 + 0,042 + 0,0282 = 0,122 dB
The combined standard uncertainty of the attenuator and cabling network uncertainty:
uc attenuation and cabling = uc2 10dB attenuatorand cable + uc2 20 dB attenuator
uc attenuation and cabling = 0,1602 + 0,1222 = 0,201 dB
6.4.5 Mismatch during measurement

Standing wave ratios involved in the power measurement:
- EUT: ρ = 0,500;
- Power sensor: VSWR ≤ 1,15 ρ = 0,070;
- 10 dB power attenuator: VSWR ≤ 1,3 ρ = 0,130;
- 20 dB attenuator: VSWR ≤ 1,25 ρ = 0,111;
- Cable: VSWR ≤ 1,2 ρ = 0,091.

The mismatch uncertainties are calculated using formula 6.3 for the individual mismatch uncertainties between:
- the standard uncertainty of the mismatch between the EUT and cable:
0,200×0,091×100%
u j mismatch:EUT tocable = = 0,112 dB
2 ×11,5
- the standard uncertainty of the mismatch between the cable and 10 dB power attenuator:
0,091×0,130×100%
u j mismatch:cableto10dB att. = = 0,073 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 10 dB power attenuator and 20 dB attenuator:
0,130×0,111×100%
u j mismatch:10dB att.to 20dB att. = = 0,089 dB
2 ×11,5
- the standard uncertainty of the mismatch between the 20 dB attenuator and power sensor:
0 ,111×0,070×100%
u j mismatch:20dB att.to power sensor = = 0,048 dB
2 ×11,5
- the standard uncertainty of the mismatch between the EUT and 10 dB power attenuator:
0 ,200×0,130×0,966 2 ×100%
u j mismatch:EUT to10dB att. = = 0,149 dB
2 ×11,5
- the standard uncertainty of the mismatch between the cable and 20 dB attenuator:
0,091×0 ,111×0,9662 ×0,3162 ×100%

u j mismatch:cableto 20dB att. = = 0,058 dB
2 × 11,5
ETSI
47 ETSI TR 100 028-1 V1.4.1 (2001-12)
- the standard uncertainty of the mismatch between the EUT and 20 dB attenuator:
0,200×0,111×0,9662 ×0,3162 ×100%

u j mismatch:EUT to 20dB att. = = 0,013 dB
2 × 11,5
The rest of the combinations:
- 10 dB attenuator to power sensor;
- cable to power sensor;
- EUT to power sensor;
are neglected. The combined standard uncertainty of the mismatch during the measurement is the RSS of the individual
components:
u c mismatch = 0 ,112 2 + 0,073 2 + 0 ,089 2 + 0,048 2 + 0,149 2 + 0,058 2 + 0,013 2 = 0,232 dB
In the case where all contributions are considered as independent.
6.4.6 Influence quantities

The two influence quantities involved in the measurement are ambient temperature and supply voltage.
Temperature uncertainty: ±1,0°C.
Supply voltage uncertainty: ±0,1 V.
Uncertainty caused by the temperature uncertainty: Dependency function (from TR 100 028-2 [8], table F.1): Mean
value 4 %/°C and standard deviation: 1,2 %/°C.
Standard uncertainty of the power uncertainty caused by ambient temperature uncertainty (formula 5.2; see also clause
D.4.2.1 of TR 100 028-2 [8]).
u j power / temperature =
1 1,02
23,0 3
( )
4,02 + 1,22 = 0,105 dB
Uncertainty caused by supply voltage uncertainty: Dependency function (from TR 100 028-2 [8], table F.1): Mean:
10 %/V and standard deviation: 3 %/V power, Standard uncertainty of the power uncertainty caused by power supply
voltage uncertainty (formula 5.2; see also clause D.4.2.1 in TR 100 028-2 [8]).
u j power / voltage =
1
23,0
0,12
3
( )
10 2 + 3 2 = 0,026 dB
uci nfluence = u 2j power / temperature + u 2j power / voltage = 0 ,1052 + 0,0262 = 0,108 dB
6.4.7 Random
The measurement was repeated 9 times The following results were obtained (before correcting for cabling and
attenuator network insertion loss):
- 21,8 mW, 22,8 mW, 23,0 mW, 22,5 mW, 22,1 mW, 22,7 mW, 21,7 mW, 22,3 mW, 22,7 mW.
The two sums X and Y are calculated:
- X = the sum of the measured values = 201,6 mW;
- Y = the sum of the squares of the measured values = 4 517,5 mW2;
ETSI
48 ETSI TR 100 028-1 V1.4.1 (2001-12)
X2 201,6 2
Y− 4517 ,5 −
- u c random = n = 9 = 0,456 mW (formula 5.5);
n −1 9 −1
- Mean value = 22,4 mW.
As the result is obtained as the mean value of 9 measurements the standard uncertainty (converted to dB by division
with 23,0) of the random uncertainty is:
0,456 100
u c random = × = 0,089 dB
22,4 23,0
NOTE: It is important to try and identify whether this value corresponds to the effect of other uncertainties,
already taken into account in the calculations (e.g. uncertainties due to the instrumentation), or whether
this corresponds to a genuine contribution (in which case it has to be combined with all the other
contributions)… obviously, there are uncertainties in the measurements, so it has to be expected that
performing the same measurement a number of times may provide a set of different results.
6.4.8 Expanded uncertainty

The combined standard uncertainty for the carrier power measurement is the RSS of all the calculated part standard
uncertainties:
uc carrier power = uc2 meter and sensor + uc2 attenuation and cabling + uc2 mismatch + uc2 in fluence + uc2 random
u ccarrier power = 0,066 2 + 0,2012 + 0 ,232 2 + 0 ,108 2 + 0,089 2 = 0,344 dB
The expanded uncertainty is ±1,96 × 0,344 dB = ±0,67 dB at a 95 % confidence level, should the distribution
corresponding to the combined uncertainty be Normal (this is further discussed in clause D.5.6 in TR 100 028-2 [8]).
The dominant part of this expanded uncertainty is mismatch uncertainty. In the calculations all the mismatch
uncertainties were based on manufacturers data, which are normally very conservative. The relevant reflection
coefficients could be measured by means of a network analyser or reflection bridge. This would probably give lower
reflection coefficients thereby reducing the overall uncertainty.
NOTE: In the case where these coefficient are measured a number of times, under conditions where it can be
considered that the measurements are independent, then the comments found in clauses 5.3.1 and 6.4.7
may be relevant.
ETSI
49 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.5 Uncertainty calculation for measurement of a receiver

(Third order intermodulation)
Before starting we need to know the architecture and the corresponding noise behaviour of the receiver.
6.5.1 Noise behaviour in different receiver configurations

The effect of noise on radio receivers is very dependant on the actual design. A radio receiver has (generally) a front
end and demodulation stages according to one of the possibilities presented in figure 13. This simplified diagram (for
AM and FM/PM systems) illustrates several possible routes from the front end to the "usable output".
FM / PM Data
Sub carrier
AM Speech
Demodulator
AM modulated data
Front end
IF
FM / PM Speech
Sub carrier
Sub carrier mod
FM / PM Data
Sub carrier
FM / PM Speech
Demodulator
Direct mod
Data
Figure 13: Possible receiver configurations
The Amplitude Modulation route involves a 1:1 conversion after the front end and the amplitude demodulation
information is available immediately (analogue) or undergoes data demodulation.
The frequency modulation/phase modulation route introduces an enhancement to the noise behaviour in non-linear
(e.g. FM/PM) systems compared to linear (e.g. AM) systems, see figure 14, until a certain threshold or lower limit
(referred to as the knee-point) is reached. Below this knee-point the demodulator output signal to noise ratio degrades
more rapidly for non-linear systems than the linear system for an equivalent degradation of the carrier to noise ratio, this
gives rise to two values for the slope: one value for C/N ratios above the knee and one value for C/N ratios below the
knee.
A similar difference will occur in data reception between systems which utilize AM and FM/PM data. Therefore "Noise
Gradient" corresponds to several entries in TR 100 028-2 [8], table F.1.
ETSI
50 ETSI TR 100 028-1 V1.4.1 (2001-12)
)
PM
/
M
(F
)
M
em
Better >
(A
yst
m
s
te
ar
ys
n e
s
-li
ar
on
ne
N
Li
S/N
Knee point
Better >
C/N
Figure 14: Noise behaviour in receivers
6.5.2 Sensitivity measurement

The sensitivity of a receiver is usually measured as the input RF signal level which produces a specific output
performance which is a function of the base band signal-to-noise ratio in the receiver.
This is done by adjusting the RF level of the input signal at the RF input of the receiver.
What is actually done is that the RF signal-to-noise ratio at the input of the receiver is adjusted to produce a specified
signal-to-noise ratio dependant behaviour at the output of the receiver, i.e. SINAD, BER, or message acceptance.
An error in the measurement of the output performance will cause a mis adjustment of the RF level and thereby the
result.
In other words any uncertainty in the output performance is converted to signal-to-noise ratio uncertainty at the input of
the receiver. As the noise does not change it causes an uncertainty in the adjusted level.
For an analogue receiver, the dependency function to transform the SINAD uncertainty to the RF input level uncertainty
is the slope of the noise function described above in clause 6.5.1 and depends on the type of carrier modulation.
The dependency function involved when measuring the sensitivity of an FM/PM receiver is the noise behaviour usually
below the knee-point for a non-linear system, in particular in the case of data equipment. This function also affects the
uncertainty when measuring sensitivity of an FM/PM based data equipment.
This dependency function has been empirically derived at 0,375 dB RF i/p level / dB SINAD associated with a standard
uncertainty of 0,075 dB RF i/p level / dB SINAD and is one of the values stated in TR 100 028-2 [8], table F.1.
If the receiver is for data the output performance is a specified BER. BER measurements are covered by clause 6.6.
In some standards the sensitivity is measured as the output performance at a specified input level. In this case the
dependency functions converting input level uncertainty to output performance uncertainty are the inverse of the
functions previously described.
ETSI
51 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.5.3 Interference immunity measurements

Interference immunity (i.e. co-channel rejection, adjacent channel rejection) is measured by adjusting the RF level of
the wanted signal to a specified value. Then the RF level of the interfering signal is adjusted to produce a specified
performance at the output of the receiver.
The interfering signal is normally modulated. Therefore for measurement uncertainty purposes it can be regarded as
white noise in the receiving channel.
The uncertainty analysis is therefore covered by clause 6.5.2.
6.5.4 Blocking and spurious response measurements

These measurements are similar to interference immunity measurements except that the unwanted signal is without
modulation.
Even though the unwanted signal (or the derived signal in the receive channel caused by the unwanted signal) can not in
every case be regarded as white noise, the present document does not distinguish. The same dependency functions are
used.
6.5.5 Third order intermodulation

When two unwanted signals X and Y occur at frequency distance d(X) and 2d(Y) from the receiving channel a disturbing
signal Z is generated in the receiving channel due to non linearities in filters, amplifiers and mixers.
The physical mechanism behind the intermodulation is the third order component of the non-linearity of the receiver:
K × X3
When two signals - X and Y - are subject to that function, the resulting function will be:
- K(X + Y)3 = K(X3 + Y3 + 3X2Y + 3XY2), where the component Z = 3X2Y is the disturbing intermodulation
product in the receiving channel.
If X is a signal Ix sin(2π(fo + d)t) and Y is a signal Iy sin(2π(fo + 2d)t), the component:
- Z = K × 3X2Y will generate a signal having the frequency fo and the amplitude K × 3Ix2Iy.
(A similar signal Z' = 3XY2 is generated on the other side of the two signals X and Y, as shown in figure 15).
The predominant function is a third order function:
Iz = Ic + 2Ix + Iy (6.4)
where lz is the level of the intermodulation product Z, Ic is a constant, Ix and Iy are the levels of X and Y. All terms are
logarithmic.
6.5.5.1 Measurement of third order intermodulation

The measurement is normally carried out as follows.
Three signal generators are connected to the input of the EUT.
Generator 1 is adjusted to a specified level at the receiving frequency fo (the wanted signal W).
Generator 2 is adjusted to frequency fo+ δ (unwanted signal X) and generator 3 is adjusted to frequency fo+ 2δ
(unwanted signal Y). The level of X and Y (Ix and Iy) are maintained equal during the measurement.
Ix and Iy are increased to level A which causes a specified degradation of AF output signal (SINAD) or a specific bit
error ratio (BER) or a specific acceptance ratio for messages.
Both the SINAD, BER and message acceptance ratio are a function of the signal-to-noise ratio in the receiving channel.
ETSI
52 ETSI TR 100 028-1 V1.4.1 (2001-12)
The level of the wanted signal W is Aw (see figure 15). The measured result is the difference between the level of the
wanted signal Aw and the level of the two unwanted signals A. This is the ideal measurement.
Level
A
X Y
A
w
W
Az
Z Z'
fo fo+δ fo+2δ
Figure 15: Third order intermodulation components
When looked upon in logarithmic terms a level change δIx dB in X will cause a level change of 2 × δIx dB in Z, and a
level change δIy dB in Y will cause the same level change δIz dB in Z.
If the levels of both X and Y are changed by δI dB, the resulting level change of Z is 3 × δI dB.
Since X is subject to a second order function, any modulation on X will be transferred with double uncertainty to Z (see
also annex D, clauses D.3.2, D.3.4 and D.5 in TR 100 028-2 [8]), whereas the deviation of any modulation on Y will be
transferred unchanged to Z.
Therefore, as Y is modulated in the measurement, the resulting modulation of Z will be the same as with Y.
6.5.5.2 Uncertainties involved in the measurement

The predominant uncertainty sources related to the measurement are the uncertainty of the levels of the applied RF
signals and uncertainty of the degradation (the SINAD, BER, or message acceptance measurement). The problems
about the degradation uncertainty are exactly the same as those involved in the co-channel rejection measurement if the
intermodulation product Z in the receiving channel is looked upon as the unwanted signal in this measurement.
Therefore the noise dependency is the same, but due to the third order function the influence on the total uncertainty is
reduced by a factor 3 (see clauses D.3.2 and D.5 in TR 100 028-2 [8]).
It is in the following assumed that the distance to the receiver noise floor is so big that the inherent receiver noise can be
disregarded.
ETSI
53 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.5.5.2.1 Signal level uncertainty of the two unwanted signals

A is the assumed level of the two unwanted signals (the indication of the two unwanted signal generators corrected for
matching network attenuations):
- Ax is the true level of X and Ay is the true level of Y. (Ax is A + δx and Ay is A + δy) see figure 16;
- Az is the level of Z (the same as in the ideal measurement).
Level
Ax
δX X
A
δY
Ay
Y
A w
W
Az
Z Z'
fo fo+δ fo+2δ
Figure 16: Level uncertainty of two unwanted signals
If Ax and Ay were known the correct measuring result would be obtained by adjusting the two unwanted signals to the
level At (true value) which still caused the level Az of Z.
If there is an error δx of the level of signal X, the error of the level of the intermodulation product will be 2 × δx (see
also clauses D.3 and D.5 in TR 100 028-2 [8]); to obtain the wanted signal-to-noise ratio the two unwanted levels must
be reduced by 2 × δx/3.
In other words the dependency function of generator X is 2/3.
In the same way if there is an error δy of the level of signal Y, the error of the level of the intermodulation product will
be δy; to obtain the wanted signal-to-noise ratio the two unwanted signals must be reduced by δy/3.
In other words the dependency function of generator Y is 1/3.
When looking at the problem in linear terms, the dependency functions are valid for small values of δx and δy due to the
fact that the higher order components of the third order function can be neglected.
δx and δy are the relative RF level uncertainties at the input of the EUT. They are combinations of signal generator level
uncertainty, matching network attenuation uncertainty and mismatch uncertainties at the inputs and the output of the
matching network.
The standard uncertainties of the levels of X and Y are uj x and uj y.
The standard uncertainty uj unwanted signals related to the uncertainty caused by level uncertainty of the two unwanted
signals is thus (see also clause D.3.2.3 of TR 100 028-2 [8]):
2 2
2  1 
u j unwanted signals =  uj x + u j y  (6.5)
3  3 
ETSI
54 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.5.5.2.2 Signal level uncertainty of the wanted signal

Under the assumption that equal change of both the level of the wanted signal and the intermodulation product will
cause no change of the SINAD, (or the BER, or the message acceptance) the error contribution from the uncertainty of
the level of the wanted signal can be calculated.
If there is an error δw on the wanted signal, the two unwanted signal levels must be adjusted by 1/3 × δw to obtain the
wanted signal-to-noise ratio. The dependency function of generator W is therefore 1/3 and assuming the same types of
uncertainties as previously the standard uncertainty, uj wanted signal, is (see clause D.3.2.3 of TR 100 028-2 [8]):
1 
u j wanted signal= u j unwanted signals  (6.6)
3 
6.5.5.3 Analogue speech (SINAD) measurement uncertainty

Sensitivity is normally stated as an RF input level in conducted measurements.
For analogue systems this is stated as at a specified SINAD value.
For an analogue receiver, the dependency function to transform the SINAD uncertainty to the RF input level uncertainty
is the slope of the noise function described above in clause 6.5.1 and depends on the type of carrier modulation.
The dependency function involved when measuring the sensitivity of an FM/PM receiver is the noise behaviour usually
below the knee-point for a non-linear system, in particular in the case of data equipment. This function also affects the
uncertainty when measuring sensitivity of an FM/PM based data equipment.
This dependency function has been empirically derived at 0,375 dB RF i/p level / dB SINAD associated with a standard
uncertainty of 0,075 dB RF i/p level / dB SINAD and is one of the values stated in TR 100 028-2 [8], table F.1.
The SINAD measurement uncertainty also contributes to the total measurement uncertainty.
If the receiver is working beyond the demodulator knee point any SINAD uncertainty corresponds to an equal
uncertainty (in dB) of the signal-to-noise ratio.
If the receiver is working below the knee point the corresponding uncertainty of the signal-to-noise ratio will be in the
order of 1/3 times the SINAD uncertainty (according to TR 100 028-2 [8], table F.1).
Any signal-to-noise ratio uncertainty causes 1/3 times that uncertainty in the combined uncertainty: the unwanted signal
levels must be adjusted by 1/3 of the signal-to-noise ratio error to obtain the correct value.
Therefore if the receiver is working above the knee point the SINAD dependency function is 1/3, and if the receiver is
working below the knee point the dependency function is in the order of 1/9.
6.5.5.4 BER and message acceptance measurement uncertainty

Any BER (or message acceptance) uncertainty will influence the total uncertainty by the inverse of the slope of the
appropriate BER function at the actual signal-to-noise ratio.
As the BER function is very steep, the resulting dependency function is small, and it is sufficient to use the differential
coefficient as an approximation.
If the signalling is on a sub carrier, the relation between the signal-to-noise ratio of the sub carrier must be dealt with in
the same way as with other receiver measurements. See clause 6.6.3.
6.5.5.5 Other methods of measuring third order intermodulation

Some test specifications specify other methods of measuring the intermodulation rejection.
The measured result is the SINAD, BER, or message acceptance at fixed test signal levels. This is the case with some
digital communication equipment like DECT and GSM.
ETSI
55 ETSI TR 100 028-1 V1.4.1 (2001-12)
In these measurements the uncertainty must be calculated in 3 steps:
1) the uncertainty of the resulting signal-to-noise ratio is calculated;
2) this uncertainty is then applied to the appropriate SINAD, BER, or message acceptance function;
3) and then combined with the measurement uncertainty of the SINAD, BER, or message acceptance measurement.
The uncertainty of the signal-to-noise ratio due to uncertainty of the level of the test signals is:
u j SNR = (2u j x )2 + u j y 2 + u 2j w
This uncertainty is then transformed to the measured parameter.
If the measured value is a SINAD value and the receiver is working beyond the knee point the SINAD uncertainty is
identical, but if the receiver is working below the knee point the dependency function is in the order of 3,0.
If the measurand is a BER or a message acceptance, the dependency function is too non linear to be regarded as a first
order function.
The total uncertainty must then be calculated as described in clause 6.6.4.3.
6.6 Uncertainty in measuring continuous bit streams

6.6.1 General
If an EUT is equipped with data facilities, the characteristic used to assess its performance is the Bit Error Ratio (BER).
The BER is the ratio of the number of bits in error to the total number of bits in a received signal and is a good measure
of receiver performance in digital radio systems just as SINAD is a good measure of receiver performance in analogue
radios. BER measurements, therefore, are used in a very similar way to SINAD measurements, particularly in
sensitivity and immunity measurements.
6.6.2 Statistics involved in the measurement

Data transmissions depend upon a received bit actually being that which was transmitted. As the level of the received
signal approaches the noise floor (and therefore the signal to noise ratio decreases), the probability of bit errors (and the
BER) increases.
The first assumption for this statistical analysis of BER measurements is that each bit received (with or without error) is
independent of all other bits received. This is a reasonable assumption for measurements on radio equipment, using
binary modulation, when measurements are carried out in steady state conditions. If, for instance, fading is introduced,
it is not a reasonable assumption.
The measurement of BER is normally carried out by comparing the received data with that which was actually
transmitted. The statistics involved in this measurement can be studied using the following population of stones: one
black and (1/BER)-1 white stones. If a stone is taken randomly from this population, its colour recorded and the stone
replaced N times, the black stone ratio can be defined as the number of occurrences of black stones divided by N. This
is equivalent to measuring BER.
The statistical distribution for this measurement is the binomial distribution. This is valid for discrete events and gives
the probability that x samples out of the N stones sampled are black stones (or x bits out of N received bits are in error)
given the BER:
× BER x (1 − BER ) N − x
N!
P( x ) = (6.7)
x!(N − x )!
ETSI
56 ETSI TR 100 028-1 V1.4.1 (2001-12)
The mean value of this distribution is BER × N and the standard deviation is:
BER × (1 − BER ) × N (6.8)
and for large values of N the shape of the distribution approximates a Gaussian distribution.
Normalizing the mean value and standard uncertainty (by dividing by N) gives:
Mean value = BER (6.9)
BER(1 − BER )
u j BER = (6.10)
N
From these two formulas it is easy to see that the larger number of bits, the smaller the random uncertainty, and the
relation between number of bits and uncertainty is the same as for random uncertainty in general. By means of
formula 6.11 it is possible to calculate the number of bits needed to be within a specific uncertainty.
For example: A BER in the region of 0,01 is to be measured.
a) If the standard uncertainty, due to the random behaviour discussed above, is to be 0,001, then the number of bits
to be compared, N, in order to fulfil this demand is calculated from the rearranged formula (6.11).
BER(1 − BER ) 0,01 × 0 ,99

N= = = 9 900
u 2jBER 0,0012
b) If the number of bits compared, N, is defined, e.g. 2 500 then the standard uncertainty is given directly by
formula (6.11).
0,01(1 − 0,01)
u j BER = = 0,002
2500
As stated earlier the binomial distribution can be approximated by a Normal distribution. This is not true when the BER
is so small that only a few bit errors (< 10) are detected within a number of bits. In this case the binomial distribution is
skewed as the p (BER < 0) = 0.
Another problem that occurs when only few bit errors are detected, and the statistical uncertainty is the dominant
uncertainty (which does not happen in PMR measurements, but it does, due to the method, occur in DECT and GSM
tests) is that the distribution of the true value about the measured value can be significantly different from an assumed
Normal distribution.
6.6.3 Calculation of uncertainty limits when the distribution characterizing

the combined standard uncertainty cannot be assumed to be a
Normal distribution
In the calculations of uncertainty there is usually no distinction between the distribution of a measured value about the
true value, and the distribution of the true value about a measured value. The assumption is that they are identical.
This is true in the cases where the standard uncertainty for the distribution of the measured value about the true value is
independent of the true value - which usually is the case. But if the standard uncertainty is a function of the true value of
the measurand (not the measured value), the resulting distribution of the measurement uncertainty will not be a Normal
distribution even if the measured value about the true value is.
This is illustrated by the following (exaggerated) example.
A DC voltage is to be measured. We assume that there is only one uncertainty contribution which comes from the
voltmeter used for the measurement.
In the manufacturers data sheet for the voltmeter it is stated that the measured value is within ±25 % of the true value.
ETSI
57 ETSI TR 100 028-1 V1.4.1 (2001-12)
If the true value is 1,00 V then the measured value lies between 0,75 V and 1,25 V. However, if the measured value is
0,75 V and the true value is still 1,00 V corresponding to 1,3333 times the measured value. Similarly, If the measured
value is 1,25 V and the true value is still 1,0 V this corresponds to 0,8 times the measured value.
Therefore the limits are asymmetric for the true value about the measured value (-20 % and +33,33 %).
When looking at the standard deviations, the error introduced is small. In the previous example the standard deviation
of the measured value about the true value is 14,43 %. The standard deviation of the related true value about the
measured value is 15,36 %. As the difference is small, and the distribution of the measured value about the true value is
based on an assumption anyway, the present document suggests that it can be used directly.
NOTE: The average value, however, is no longer zero, but in this case is approximately 4,4 %.
Alternatively, also in this example, xt is the true value and xm is the measured value. Any parameter printed in square
brackets, e.g. [xm], is considered to be constant.
The distribution of the measured value xm about the true value xt is given by the function p (xm, [xt]).
Based on this function the distribution p1 (xt, [xm]) of the true value xt about the measured value xm can be derived.
The intermediate function is p (xt, [xm]) which is the same as the previous; the only difference being that xt is the
variable and xm is held constant. This function is not a probability distribution as the integral from -∞ to +∞ is not unity.
To be converted to the probability function p1 (xt, [xm]) it must be normalized. Therefore:
p(x t ,[x m ])
p1( xt ,[x m ]) = (6.11)
∞
∫ p(x,[x
−∞
m ])dx
As this distribution is not Normal, the uncertainty limits must be found by other means than by multiplication with a
coverage factor from Student's t-distribution. How the actual limits are calculated in practise depends on the actual
distribution.
An example: If the true BER of a radio is 5 × 10-6 and the BER is measured over 106 bits, the probability of detecting
0 bits is 0,674 %. On the other hand if the BER in a measurement is measured as 5 × 10-6 the true value cannot be 0.
If the uncertainty calculations are based on the assumption of a Gaussian distribution, the lower uncertainty limit
becomes negative (which of course does not reflect reality, and provides the evidence that not all distributions are
Normal!).
The standard uncertainty based on the measured value 3,0 × 10-6:
uj =
(
3,0×10 −6 1 − 3,0×10 −6 ) = 1,73 ×10 −6
6
10
The expanded uncertainty is ±1,96 × 1,73 × 10-6 = ±3,39 × 10-6 at a 95 % confidence level.
The correct distribution p1 (xt) is the continuous function in figure 17.
NOTE: The true value is not BER, but number of bit errors, where BER= (bit errors/number of bits tested)). The
binomial function p (xm) based on the true value = 3 bit errors (corresponding to BER = 3 × 10-6) is the
discrete function shown.
The distribution p (xt) (based on the binomial distribution with 3 bit errors and 106 bits tested):
 x 
3
 x 
(10 −3)
6
10 6 !
p( )
x t = 100×k × t × 1 − t % where k =
( ) = 1,67×10
 
17
 
 10
6
  10 6  3!× 10 6 − 3 !
ETSI
58 ETSI TR 100 028-1 V1.4.1 (2001-12)
The integral from -∞ to +∞ of p (xt) is very close to 1. Therefore p (xt) is a good approximation to the correct
distribution p1 (xt).
By means of numerical methods the 95 % error limits are found to be +5,73 and -1,91 corresponding to +5,73 × 10-6
and -1,91 × 10-6.
Figure 17 shows the discrete distribution giving the probabilities of measuring from 0 to 14 bit errors when the true
value is 3 bit errors corresponding to BER = 3 × 10-6, and the continuous distribution giving the probability function for
the true value when the measured value is 3 bit errors corresponding to BER = 3 × 10-6.
25
p1 (x t )
20
15
%
10
5
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
bit errors
Figure 17: BER uncertainty
6.6.4 BER dependency functions

As in SINAD measurements, the BER of a receiver is a function of the signal to noise ratio of the RF signal at the input
of the receiver.
Several modulation and demodulation techniques are used in data communication and the dependency functions are
related to these techniques.
This clause covers the following types of modulation:
- coherent modulation/demodulation of the RF signal;
- non coherent modulation/demodulation of the RF signal;
- FM modulation.
The following assumes throughout that the data modulation uncertainty combines linearly to the carrier to noise ratio
uncertainty. The uncertainty calculations are based on ideal receivers and demodulators where correctly matched filters
are utilized.
ETSI
59 ETSI TR 100 028-1 V1.4.1 (2001-12)
The characteristics of practical implementations may differ from the theoretical models thereby having BER
dependency functions which are different from the theoretical ones. The actual dependency functions can, of course, be
estimated individually for each implementation. This, however, would mean additional measurements. Instead the
theoretically deduced dependency functions may be used in uncertainty calculations.
6.6.4.1 Coherent data communications

Coherent demodulation techniques are techniques which use absolute phase as part of the information. Therefore the
receiver must be able to retrieve the absolute phase from the received signal. This involves very stable oscillators and
sophisticated demodulation circuitry, but there is a gain in performance under noise conditions compared to non
coherent data communication. Coherent demodulation is used, for example, in the GSM system with Gaussian
Minimum Shift Keying (GMSK).
6.6.4.2 Coherent data communications (direct modulation)

The BER as a function of SNRb, the signal to noise ratio per bit for coherent binary systems is:
BER (SNRb) = 0,5 × erfc (√SNRb) (6.12)
where erfc (x) is defined as:

∞
∫e
2 -t 2
erfc(x)= dt (6.13)
π
x
It is not possible to calculate the integral part of (6.11) analytically, but the BER as a function of the signal to noise ratio
is shown in figure 18 together with the function for non coherent binary data communication.
There are different types of coherent modulation and the noise dependency of each varies, but the shape of the function
remains the same. The slope, however, is easily calculated and, although it is negative, the sign has no meaning for the
following uncertainty calculations:
d (BER ) 1
= × e − SNRb (6.14)
d (SNRb ) 2 π × SNRb
For the purpose of calculating the measurement uncertainty, this can be approximated:
d (BER )
≈ 1,2×BER (6.15)
d (SNRb )
If the aim is to transform BER uncertainty to level uncertainty - which is the most likely case in PMR measurements,
the inverse dependency function must be used (the result is in percentage power terms as it is normalized by division
with SNRb*):
 
 u jBER  u jBER
u j leveldueto BERuncertainty =  100% ≈ ×100% (6.16)
(
 d BER )
× SNRb*
 1,2×BER×SNRb*

(
 d SNRb ) 

The SNRb* is a theoretical signal to noise ratio read from figure 19. It may not be the signal to noise ratio at the input of
the receiver but the slope of the function is assumed to be correct for the BER measured.
For example: The sensitivity of a receiver is measured. The RF input level to the receiver is adjusted to obtain a BER of
10-2. The measured result is the RF level giving this BER. The BER is measured over a series of 25 000 bits. The
resulting BER uncertainty is then calculated using formula (6.11):
0,01(1 − 0 ,01)
u j BER = = 6,29 × 10 − 4
25000
ETSI
60 ETSI TR 100 028-1 V1.4.1 (2001-12)
The uncertainty of the RF signal at the input is 0,7 dB (uj). The signal to noise ratio giving this BER is then read from
figure 18: SNRb*(0,01) = 2,7 and the dependency function at this level is:
d (BER(2,7 ))
= 1,2 × BER = 1,2 × 1 × 10 −2 = 1,2 × 10 − 2
d (SNRb )
The BER uncertainty is then transformed to level uncertainty using formula (6.16):
 6 ,29 × 10 −4  1,95
u j level =   × 100 % = 1,95 % power ≈ dB = 0,085 dB
−2
1,2×10 ×
 2,7  23,0
u j RF level = 0,7 2 + 0 ,085 2 + ....
There is an additional uncertainty component due to resolution of the readout of the measured BER. If the RF input
level has been adjusted to give a reading of 0,01 and the resolution of the BER meter is 0,001 the correct lies between
0,0095 and 0,0105 with equal probability.
The standard deviation is therefore
0,5 × 10−3
u j BER resolution = = 2,89 × 10−4
3
This standard deviation is then by means of formula 6.16 converted to level uncertainty:
u j BER resolution
u j level due to BER resolution = × 100 %
d(BER)
× SNRb
d(SNRb )
0 ,289 × 10 −4 0 ,089
u j level due to BER resolution = × 100 % = 0,089 % ≈ dB = 0,004 dB
1,2 × 0,01 × 2,7 23,0
The total uncertainty of the sensitivity level is then:
uc RF level = u 2j levelu 2j level dueto BER resolution + u 2j = 0,0852 + 0,0042 + 0,72 = 0,71dB
As can be seen the BER statistical uncertainty and the BER resolution only plays a minor role.
6.6.4.3 Coherent data communications (sub carrier modulation)

If a sub carrier frequency modulation is used in the data communication the functions related to direct coherent data
communication apply, but in this case they give the relationship between BER and the signal to noise of the sub carrier.
To be able to transform BER uncertainty to RF input level uncertainty the relationship between the sub carrier signal to
noise ratio and the RF carrier signal to noise ratio must be calculated.
If the BER is measured at a RF level much higher than the sensitivity this relation is assumed to be 1:1 as described in
clause 6.5.
In FM systems, if the BER is measured in the sensitivity region (below the knee point) the relationship as for analogue
receivers is assumed and the same value taken from TR 100 028-2 [8], table F.1, 0,375 dB RF i/p level/dB SINAD and
standard uncertainty 0,075 dB RF i/p level/dB SINAD. (see clause 6.5).
EXAMPLE: The sensitivity of an FM receiver is measured. The RF input level to the receiver is adjusted to
obtain a BER of 10-2. The measured result is the RF level giving this BER. The BER is measured
over a series of 2 500 bits. The uncertainty of the RF signal at the input is 0,5 dB (uj).
ETSI
61 ETSI TR 100 028-1 V1.4.1 (2001-12)
The resulting BER uncertainty is then calculated using formula (6.11):
0 ,01(1 − 0,01)
u j BER = = 2,0 × 10 −3
2500
The signal to noise ratio giving this BER is then read from figure 18: SNRb*(0,01) = 2,7. The dependency function at
this level is:
d (BER(2 ,7 ))
= 1,2×BER = 1,2×1,0×10 −2 = 1,2×10 − 2
d (SNRb )
The BER uncertainty is then transformed to level (or SNRb) uncertainty using formula (6.4):
 2,0×10 −3 
u j SNRb =  ×100 % = 6,17 % power, which is equal to (6,17/23,0) = 0,27 dB.
−2
1,2×10 ×2 ,7 
 
This uncertainty is then by means of formula (5.2) and the relationship taken from TR 100 028-2 [8], table F.1
converted to RF input level uncertainty (as SINAD and SNRb is considered to be equivalent in this case). The
dependency function is: mean = 0,375 dB RF i/p level/dB SINAD and standard uncertainty 0,075 dB RF i/p level/dB SINAD.
( )
u j level = 0,27 2 × 0,38 2 + 0,08 2 = 0 ,102 dB (formula 5.2)
This RF level uncertainty is then combined with the uncertainty of the level of the input signal to obtain the total
uncertainty of the sensitivity:
u j sensitivity = 0,5 2 + 0,10 2 = 0,51 dB
In this example the uncertainty due to meter resolution is assumed to be negligible.
6.6.4.4 Non coherent data communication

Non coherent modulation techniques disregard absolute phase information. Communications based on non coherent
modulation tend to be more sensitive to noise, and the techniques used may be much simpler. A typical non coherent
demodulation technique is used with FSK, where only the information of the frequency of the signal is required.
6.6.4.5 Non coherent data communications (direct modulation)

The BER as a function of the SNRb in this case is:
SNRb
1 −
BER(SNRb ) = e 2 (6.17)
2
provided that the cross correlation coefficient ccross between the two frequencies defining the zeros and the ones is 0.
The cross correlation coefficient ccross of two FSK signals with frequency separation fδ and the bit time T is:
sin (π × T × f δ )
c cross = (6.18)
π × T × fδ
It is assumed that the cross correlation coefficient for land mobile radio systems is so small that the formulas for
ccross = 0 apply, and as ccross is 0 the BER, as a function of the SNRb for non coherent modulation is shown in
equation 6.15.
The slope of the function (in fact the slope is negative, but the sign is of no interest for the uncertainty calculation). The
BER (SNRb) function for non coherent data communication is shown in figure 19.
ETSI
62 ETSI TR 100 028-1 V1.4.1 (2001-12)
The inverse function is:
SNRb (BER) = -2 × ln (2 × BER) (6.19)
From (6.17) the slope of SNRb (BER) is:
d (SNRb ) 2
=− (6.20)
d (BER ) BER
The slope of the function is the inverse of (6.18):
d (BER ) BER
= (6.21)
d (SNRb ) 2
The SNRb can be calculated by means of formula (6.19) or read from the function shown in figure 19. If the aim is to
transform BER uncertainty to level uncertainty - which is generally the case in PMR measurements - formula (6.16) is
used.
u j BER
u j level =
d (BER )
×SNRb*
d (SNRb )
Before it can be combined with the other part uncertainties at the input of the receiver it must be transformed to linear
voltage terms.
EXAMPLE: The sensitivity of a receiver is measured. The RF input level to the receiver is adjusted to obtain a
BER of 10-2. The measured result is the RF level giving this BER. The BER is measured over a
series of 2 500 bits. The uncertainty of the RF signal at the input is 0,6 dB (uj).
The resulting BER uncertainty is then calculated using formula (6.11):
0 ,01× 0,99
u j BER = = 2,00 ×10 −3
2500
The signal to noise ratio giving this BER is then calculated using formula (6.19).
SNRb (0,01) = -2 × ln (2 × 0,01) = 7,824
The dependency function at this level is (formula (6.21)):
d (BER(7 ,824))
= 0,5× 0,01
d (SNRb )
The BER uncertainty is then transformed to level uncertainty using formula (6.16):
 2,00×10 −3 
u j level =  ×100 % = 5,11 % power
−2
 0 ,5 × 10
 ×7 ,824 
which is equal to 5,11/23,0 dB = 0,22 dB (uj) in voltage terms. This RF level uncertainty is then combined with the rest
of the uncertainty contribution to give the combined standard uncertainty of the RF level.
u c RF level = (0,6)2 + (0,22)2 = 0,64 dB
ETSI
63 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.6.4.6 Non coherent data communications (sub carrier modulation)

If a sub carrier modulation is used in the data communication the functions related to direct non coherent data
communications apply, but in this case they give the relation between BER and signal to noise ratio of the sub carrier.
To be able to transform BER uncertainty to RF input level uncertainty the relationship between the sub carrier signal to
noise ratio and the RF carrier signal to noise ratio must be calculated. If the BER is measured at a RF level much higher
than the sensitivity this relationship is assumed to be 1:1 as described in clause 6.5.
In FM systems, If the BER is measured in the sensitivity region (below the knee point) the relationship as for analogue
receivers is assumed and the same value taken from TR 100 028-2 [8], table F.1, 0,375 dBRF i/p level/dBSINAD and
standard uncertainty 0,075 dB RF i/p level/dB SINAD (see clause 6.5).
EXAMPLE: The sensitivity of an FM receiver is measured. The RF input level to the receiver is adjusted to
obtain a BER of 10-2. The measured result is the RF level giving this BER. The BER is measured
over a series of 2 500 bits. The uncertainty of the RF signal at the input is 0,6 dB (uj). The
0,01×0,99
u j BER = = 2,00 ×10 −3
2500
The signal to noise ratio giving this BER is then calculated using formula (6.19).
SNRb*(0,01) = -2 × ln (2 × 0,01) = 7,824
The dependency function at this level is:
d (BER(7 ,824)) 0 ,01

=
d (SNRb ) 2
This BER uncertainty is then transformed to level uncertainty using formula (6.16):
 2,00×10 −3 
u j level =  ×100 % = 5,11 % power
−2
 0 ,5×10 ×7 ,824 

which is equal to 5,11/23,0 = 0,22 dB (uj level). This sub carrier level uncertainty is then transformed to RF level
uncertainty.
u j RF leveltransformed = (0,22)2 ×((0,375dBRF /dBSINAD )2 + (0,075dBRF /dBSINAD )2 ) = 0,08 dB

NOTE: As the uncertainty is small the dependency function can be used directly without transforming to dB.
This RF level uncertainty is then combined with the uncertainty of the level of the input signal to obtain the total
uncertainty of the sensitivity:
u j sensitivity = (0,6)2 + (0,08)2 = 0,61 dB
The uncertainty due to meter resolution is assumed to be negligible.
6.6.5 Effect of BER on the RF level uncertainty

The SNRb to BER function is used to transform BER uncertainty to RF input level uncertainty. In the measurements on
PMR equipment the RF input level is adjusted to obtain a specified BER. A sufficiently large number of bits are
examined to measure the BER, but still there is a (small) measurement uncertainty contribution uj BER.
ETSI
64 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.6.5.1 BER at a specified RF level

If the purpose is to measure the BER at a specific input level, the transformation is more of a problem. The BER
function is so non-linear that the approximation where (dBER)/(dSNRB) is used as the dependency function is no longer
sufficient.
One approach is to calculate the uncertainty limits of the RF input level at the wanted confidence level, and then apply
these limits directly to the BER function. In this case the statistical uncertainty in the BER measurement is ignored, but
as the following example shows, the uncertainty due to this is negligible.
For example: The BER of a receiver is measured with the RF input level adjusted to the sensitivity limit. A BER of
0,75 × 10-2 is measured over a series of 25 000 bits. The uncertainty of the RF signal at the input is 1,1 dB (uj). The
0 ,0075 (1 − 0 ,0075)
u j BER = = 5,45×10 − 4 corresponding to 7,3 %
25000
The straight forward procedure of calculating the combined standard uncertainty by applying a 1st order dependency
function to the standard uncertainty of the RF input level uncertainty does not reflect reality due to the non linearity of
the BER function. This is shown in the following calculation.
The dependency function is 1,2 × 0,75 × 10-2 = 0,9 × 10-2 found by formula 6.3. The SNRb at BER= 0,0075 is read to
be 2,9 from figure 18. The level uncertainty of 1,1 dB corresponds to 1,1 × 23,0 % (p) = 25,5 % (uj).This is transformed
to SNRb uncertainty: 0,255 × 2,9 = 0,74 (uj). The level uncertainty is then transformed to BER uncertainty by means of
the dependency function.
uj BER = 0,74 × 0,9 × 10-2 = 0,666 × 10-2.
The expanded uncertainty = ±1,96 × 0,666 × 10-2 = ±1,31 × 10-2 at a 95 % confidence level. This expanded uncertainty
would give a negative bit error ratio as the lower limit. The reason is the non-linearity of the BER function (see also the
discussion on confidence levels and their relations with the actual distributions, in clause D.5.6.2 of TR 100 028-2 [8]).
Therefore another method should be applied.
The expanded uncertainty should be expressed at a 95 % confidence level. Therefore the input level uncertainty limits
are found to be ±1,96 × 1,1 dB = ±2,16 dB. This corresponds to 1,64 and 0,608 (power values). The values
corresponding to the 95 % confidence level is then 2,9 × 1,64 = 4,76 and 2,9 × 0,608 = 1,76.
By means of figure 18 the BER uncertainty limits at 95 % confidence level are read to be 3,0 × 10-2 and 1,0 × 10-3
corresponding to +300 % and -87 %.
ETSI
65 ETSI TR 100 028-1 V1.4.1 (2001-12)
-1
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-2
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
BER(SNRb) for coherent modulation
-3
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-4
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-5
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-6
10
1 2 3 4 5 6 7 8 9 10 11 12
Signal to noise ratio per bit SNRb (in linear terms)
Figure 18: BER (SNRb) against SNRb
ETSI
66 ETSI TR 100 028-1 V1.4.1 (2001-12)
-1
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-2
10
8,0
6,0
5,0
4,0
3,0
2,0
BER(SNRb) for noncoherent modulation
1,5
-3
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-4
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-5
10
8,0
6,0
5,0
4,0
3,0
2,0
1,5
-6
10
4 6 8 10 12 14 16 18 20 22 24 26
Signal to noise ratio per bit SNRb (in linear terms)
Figure 19: BER (SNRb) against SNRb
ETSI
67 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.6.6 Limitations in the applicability of BER uncertainty calculations

As mentioned earlier the above figures and formulas are not applicable to all BER measurements; the conditions for
applicability :
- the noise is white Gaussian noise;
- the signal-to-noise ratio is constant;
- each bit error is statistically independent;
- the transmission channel delay is constant.
These 4 conditions apply to most normal receiver measurements covered by this document, but the blocking
measurement (and any variant where the unwanted signal is un-modulated) does not satisfy the first condition about
white Gaussian noise. Therefore the formulas do not apply to this measurement.
The receiver is normally not as sensitive to a single frequency component as to a broad band signal with the same
power.
In some technologies (for instance GSM) data are protected by error correcting signalling schemes. The data are
usually transmitted in packets with extra information for the error correction attached to the packet, so that up to a
specified number of bit errors within a packet can be corrected. When this limit is exceeded the number of bits will
increase dramatically because the error correction procedures will generate more bit errors than actually received. The
result is that the BER will be less sensitive to noise at moderate signal-to-noise ratios, but the dependency function will
be steeper at lower signal-to-noise ratios. The reception of the data packet also relies on the recognition of the packet's
preamble or synchronization pattern. If this is not received and accepted all data are lost.
The dependency function depends very much on the error correction algorithm and must be analysed and derived in
each case.
In some technologies receiver characteristics are measured under fading and multi-path conditions which means that the
signal-to-noise ratio is not constant, but the multi-path conditions add other errors like distortion and timing errors of
the demodulated signal. Also it causes the bit errors to appear as bursts rather than independent errors.
In all the cases above the BER dependency functions derived previously do not apply, as one or more of the conditions
are not fulfilled.
The dependency functions must then be derived or estimated by other means.
A simple approach is to estimate the dependency function by measuring the BER at different signal-to-noise ratios, for
instance by changing the level of the wanted signal 1 dB up and down.
The problem is that the dependency functions estimated for one receiver do not necessarily apply to the next receiver
even within the same technology.
6.7 Uncertainty in measuring messages

6.7.1 General
If the EUT is equipped with message facilities the characteristic used to assess the performance of the equipment is the
Message acceptance ratio. The Message acceptance ratio is the ratio of the number of Messages accepted to the total
number of message sent.
Normally it is required to assess the receiver performance at a Message acceptance ratio of 80 %. The Message
acceptance ratio is used as a measure of receiver performance in digital radio systems in a similar way that SINAD and
BER ratios are used as a measure of receiver performance in analogue and bit stream measurements, particularly in
sensitivity and immunity measurements.
ETSI
68 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.7.2 Statistics involved in the measurement

When considering messages, parameters such as message length (in bits), type of modulation (direct or sub-carrier,
coherent or non-coherent), affect the statistics that describe the behaviour of the receiver system.
Performance of the receiver is assessed against a message acceptance ratio set by the appropriate standard and/or
methodology used. To assess the uncertainty the cumulative probability distribution curves for message acceptance are
required, these can be calculated from (6.20).
Pe (0) + Pe (1) + Pe (2) + Pe (3) + ... + Pe (n) (6.22)
Where: n is the message length:
- Pe (0) is the probability of 0 errors;
- Pe (n) is the probability of n errors.
The individual contribution of each probability Pe (x) in formula (6.22) is calculated using formula (6.8). Curves for a
theoretical 50 bit system with 1, 2, 3, 4, 5 and 6 bits of error correction are shown in figure 20.
As the number of bits of error correction increase so does the slope of the relevant portion of the cumulative probability
density function, and as the slope increases less carrier to noise (or RF input level) variation is required to cause the
message acceptance ratio to vary between 0 % and 100 %.
This effect is increased in non-linear systems by a factor of approximately 3:1. Due to the increased slope associated
with sub-carrier modulation, as a result of this in our theoretical 50 bit system, 6 bits of error correction will result in a
very well defined level of 0 % acceptance to 100 % acceptance, (with 1 dB level variation), however, with no error
correction, the level variation between 0 % and 100 % acceptance will be several dB.
Cumulative probability dist ribution

100
90
80
Probability(%)
70
60
50
40
30
20
10
0,0
2 3 4 5 6 7 8 9 10 11 12
level (dB)
Figure 20: Cumulative Probability (error correction for Messages)
ETSI
69 ETSI TR 100 028-1 V1.4.1 (2001-12)
As a method of testing receivers the "up-down" method is used. The usage of the up down method will result in a series
of transmissions using a limited number of RF levels.
6.7.3 Analysis of the situation where the up down method results in a shift
between two levels
With some systems (e.g. 6 bits of error correction) the up-down method will typically result in a pattern shifting
between two levels, where at the lower level the message acceptance ratio will approach zero and at the higher level
(+1 dB) the message acceptance ratio will approach 100 %. In this case the measurement uncertainty is of the simplest
form for this contribution.
The RF is switching between two levels, the mean value is calculated, usually from 10 or 11 measurements. The
measurement uncertainty cannot be calculated as though random, independent sources are involved. The RF is
switching between two output levels of the same signal generator, the levels therefore are correlated and only have two
values (upper and lower), hence the standard uncertainty for a signal generator with output level uncertainty of ±1 dB
is:
1,0
u j outputlevel = = 0,58 dB
3
Also there is a quantization uncertainty associated with half of the step size (in this case 1 dB which gives ±0,5 dB).
0,5
u j quantisiation = = 0,29 dB
3
Therefore the combined standard uncertainty of this step will be:
uctwolevel shift = u 2j outputlevel + u 2j quantisation = 0,582 + 0,292 = 0,65 dB
For the case of no error correction the pattern of the measured results will spread beyond a single dB step and
measurement uncertainty calculations are more complex.
6.7.4 Detailed example of uncertainty in measuring messages

For this example a theoretical system with 50 bit message length and 1 bit error correction will be considered, although
the principles can be applied to all practicable message and correction lengths.
a) Calculate the message acceptance ratio (formula (6.22)) for the given message length and given number of bit
error corrections, using bit error ratios corresponding to a convenient step size (in this case 1 dB) using either
formula (6.18) for non-coherent, or, formula (6.12) for coherent, and if sub-carrier modulation is used, use the
appropriate SINAD conversion in TR 100 028-2 [8], table F.1.
b) Now the probability of being at a given point on the curve must be assessed. For example the probability of
being at a particular point (in figure 20) is:
- the probability of being below a particular point times the probability of going up from this point; plus
- the probability of being above a particular point times the probability of going down from this point.
The method requires three successful responses, therefore the probability of going up is:
Pp (up) = 1 - (Message Acceptance)3 = 1- (MA)3 (6.23)
and the probability of going down is:
Pp (down) = (Message Acceptance)3 = (MA)3 (6.24)
(Pe (0) + Pe (1)) = Probability of 0 errors + the probability of 1 error (see formula (6.24)). These calculations are shown
in table 2.
ETSI
70 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 2: Probability of going up or down from a given position

3 3
dB Linear BER (Pe(0) + Pe(1))% Pp (up)=1- (MA) Pp (down)= (MA)
0,8826 × 10 2,698 × 10 997,3 × 10
-3 -3 -3
+2 12,679 99,91
3,251 × 10 34,69 × 10 965,3 × 10
-3 -3 -3
+1 10,071 98,83
9,158 × 10 213,7 × 10 786,3 × 10
-3 -3 -3
0 8,000 92,30
20,84 × 10 626,4 × 10 373,6 × 10
-3 -3 -3
-1 6,355 72,02
40,07 × 10 936,2 × 10 63,76 × 10
-3 -3 -3
-2 5,048 39,95
67,33 × 10 997,2 × 10 2,821 × 10
-3 -3 -3
-3 4,010 14,13
101,7 × 10 30,46 × 10
-3 -6
-4 3,185 3,123 1,000
141,1 × 10 96,55 × 10
-3 -9
-5 2,530 0,459 1,000
Based on equations (6.21) and (6.22), and the fact that the sum of all probabilities equals 1, the individual probabilities
of being at each step of the signal to noise ratio per bit (SNRb) can be calculated.
Assuming that at SNRb greater than +1 dB all messages are accepted (therefore can only move down from here) and
Assuming that at SNRb less than -4 dB all messages are rejected (therefore can only move up from here), this gives rise
to two boundary positions -5 dB and +2 dB.
The probability of being at any one of the points -5, -4, -3, -2, -1, 0, +1, +2 is Pp-5, Pp-4, Pp-3, Pp-2, Pp-1, Pp0, Pp + 1,
and Pp + 2 respectively.
The analysis of the possible transitions between these points provide:
- Pp-5 = (Pp-4 + 30,46 × 10-6) + (Pp-6 × 1);
- Pp-4 = (Pp-3 × 2,821 × 10-3) + (Pp-5 × 1);
- Pp-3 = (Pp-2 × 63,76 × 10-3) + (Pp-4 × 1);
- Pp-2 = (Pp-1 × 373,6 × 10-3) + (Pp-3 × 997,2 × 10-3);
- Pp-1 = (Pp0 × 786,3 × 10-3) + (Pp-2 × 936,2 × 10-3);
- Pp0 = (Pp + 1 × 965,3 × 10-3) + (Pp-1 × 626,4 × 10-3);
- Pp + 1 = (Pp + 2 × 1) + (Pp0 × 213,7 × 10-3);
- Pp + 2 = (Pp + 3 × 1) + (Pp + 1 × 34,69 × 10-3).
NOTE: The probability of being at point Pp-6 or Pp + 3 is zero, hence Pp-6 × 1 and Pp + 3 × 1 are both equal to
zero.
Based on seven out of these eight equations and the fact that the sum of Pp-5 to Pp + 2 is one, each individual probability
Pp-5 to Pp + 2 is calculated as follows.
Rearranging the above equations gives:
- Pp-6 × 1 - Pp-5 + Pp-4 × 30,46 × 10-6 = 0;
- Pp-5 × 1 - Pp-4 + Pp-3 × 2,821 × 10-3 = 0;
- Pp-4 × 1 - Pp-3 + Pp-2 × 63,76 × 10-3 = 0;
- Pp-3 × 997,3 × 10-3 - Pp-2 + Pp-1 × 373,6 × 10-3 = 0;
- Pp-2 × 936,2 × 10-3 - Pp-1 + Pp0 × 786,3 × 10-3 = 0;
- Pp-1 × 626,4 × 10-3 - Pp0 + Pp + 1 × 965,3 × 10-3 = 0;
ETSI
71 ETSI TR 100 028-1 V1.4.1 (2001-12)
- Pp0 × 213,7 × 10-3 - Pp + 1 + Pp + 2 × 1 = 0;
- Pp + 1 × 34,69 × 10-3 - Pp + 2 + Pp + 3 × 1 = 0;
- Pp-5 + Pp-4 + Pp-3 + Pp-2 + Pp-1 + Pp0 + Pp + 1 + Pp + 2 = 1;
- Pp-6 = Pp + 3 = 0.
Pp-5 Pp-4 Pp-3 Pp-2 Pp-1 Pp0 Pp + 1 Pp + 2

2,821 × 10
-3
1 1 -1
63,76 × 10
-3
2 1 -1
997,3 × 10 373,6 × 10
-3 -3
3 -1
936,2 × 10 786,3 × 10
-3 -3
4 -1
626,4 × 10 965,3 × 10
-3 -3
5 -1
213,7 × 10
-3
6 -1 1
34,69 × 10
-3
7 -1
8 1 1 1 1 1 1 1 1 1
Solving this by means of row operations on row 8, gives:
2,821 × 10
-3
1 1 -1
63,76 × 10
-3
2 1 -1
997,3 × 10 373,6 × 10
-3 -3
3 -1
936,2 × 10 786,3 × 10
-3 -3
4 -1
626,4 × 10 965,3 × 10
-3 -3
5 -1
213,7 × 10
-3
6 -1 1
34,69 × 10
-3
7 -1
8 392,91 1
From this we have: 392,91 × Pp + 2 = 1; therefore Pp + 2 = 2,545 × 10-3:
2,545 × 10 −3
- this is then used in row 7 to determine Pp + 1: Pp +1 = = 73,36 × 10 −3 ;
34,69 × 10 −3
- this is used in row 6 to determine Pp0: Pp 0 =

(
0,07336 − 2,545 × 10 −3 ×1 ) = 331,38 × 10 −3
;
−3
213,7 × 10
- this is used in row 5 to determine Pp-1: Pp −1 =

(
331,38 × 10 −3 − 73,36 × 10 −3 × 965,3 × 10 −3 ) = 415,97 × 10 −3
;
−3
626,4 × 10
415,97 × 10 −3 − (0 ,33138 × 0,7863)

- this is used in row 4 to determine Pp-2: Pp − 2 = = 166,0 × 10 −3 ;
0,9362
166,00 × 10 −3 − (0 ,41597 × 0,3736)

- this is used in row 3 to determine Pp-3: Pp −3 = = 10,622 × 10 −3 ;
0,9973
- this is used in row 2 to determine Pp-4: Pp − 4 =

(
10,622 × 10 −3 − 0,1660 × 63,76 × 10 −3 )
= 37 ,84 × 10 −6 ;
1
- this is used in row 1 to determine Pp-5: Pp −5 =

(
37 ,84 × 10 −6 − 10,622 × 10 −3 × 2,821 × 10 −3
= 7 ,87 × 10 −6 .
)
1
There are, off course, other ways of solving the equations.
ETSI
72 ETSI TR 100 028-1 V1.4.1 (2001-12)
After having calculated the probabilities it should always be checked that the sum of all probabilities is 1. If the sum is
not 1 (to within 0,001) it can cause major uncertainties in the calculation of the resulting standard uncertainty of the
distribution.
Based on these probabilities the standard uncertainty of the distribution is calculated:

i =2
X = ∑
=−
i 5
Ppi ×i = −0 ,70 (formula 5.6)
i=2
Y= ∑=−
i 5
Ppi ×i 2 = 1,26 (formula 5.7)
then:
u j = Y − X 2 = 1,26 − (− 0,70)2 = 0,88 dB (formula 5.8)
and the standard uncertainty for the measurement (as the result is the average value of 10 samples):
0,88
= 0,28 dB (formula 5.9)
10
Therefore the methodology introduces an additional ±0,54 dB of uncertainty to the level.
6.8 Examples of measurement uncertainty analysis (Free Field

Test Sites)
6.8.1 Introduction
This clause contains detailed worked examples of the calculation of expanded uncertainty of the radiated tests on Free
Field Test Sites i.e. Anechoic Chambers, Anechoic Chamber with a ground plane and Open Area Test Sites.
The example tests given are limited to:
- a verification procedure;
- the measurement of a transmitter parameter (spurious emission);
- the measurement of a receiver parameter (sensitivity).
All the example tests are assumed to have been carried out in an Anechoic Chamber with a ground plane since this type
of test site will contribute virtually all the uncertainty contributions involved in radiated tests on any Free Field Test Site
(i.e. all uncertainties associated with absorbing materials and ground planes).
NOTE 1: The values given to the uncertainty components in these examples are only to illustrate the uncertainty
calculations. In practice, values should be derived by consulting annex A.
All radiated tests consist of two stages. For a verification procedure, the first stage is to set a reference level whilst the
second stage involves the measurement of path loss between two antennas. For the measurement of a particular
parameter from an EUT, the first stage is usually to measure the relevant parameter whilst the second stage compares
this measurement against that from a known standard.
Within any radiated test there are uncertainty components that are common to both stages of the test. By their different
natures some of these common uncertainties will cancel (e.g. the uncertainty of the insertion loss of a cable common to
both parts) whilst others may contribute twice (e.g. the level stability of a signal generator in receiver tests). In each of
the following uncertainty analyses, uncertainty components common to both stages are shown as shaded areas in the
accompanying schematic diagrams.
ETSI
73 ETSI TR 100 028-1 V1.4.1 (2001-12)
As shown in the examples, all the individual uncertainty components for any test are combined in the manner described
in TR 102 273 [3], part 1, sub-part 1, clauses 4 and 5 in order to derive an expanded uncertainty figure for the
measurement. The values of the individual components are either provided in annex A or should be taken from
manufacturers' data sheets. Whenever they are obtained from data sheets, worst case figures given over a frequency
band should be used. For example, in the case of a signal generator whose absolute level accuracy is quoted as ±1 dB
over 30 MHz to 300 MHz, ±2 dB over 300 MHz to 1 000 MHz the figure for the band containing the test frequency
should be used. This approach should be adopted for all uncertainty components, taking the uncertainty figures over as
broad a band a possible. This is normally satisfactory when the variation with frequency is not large and provides a
simple and flexible approach. The resulting expanded uncertainty figure is valid across a broad range of frequencies and
measurement conditions and avoids the necessity of repeated calculation for minor frequency changes.
NOTE 2: Taking specific frequency values may result in a lower expanded uncertainty value, but this lower value is
only valid when that specific set of circumstances apply for which the value was derived.
6.8.2 Example 1: Verification procedure

The verification procedure is a process carried out to prove a facility's suitability as a Free Field Test Site. It involves
the transmission of a known signal level from one calibrated antenna (usually a dipole) and the measurement of the
received signal level in a second calibrated antenna (also usually a dipole). By comparison of the transmitted and
received signal levels, an "insertion loss" can be deduced.
For the measurement of NSA two stages are involved. The first is a direct attenuation measurement (Vdirect) whilst the
second is a radiated attenuation measurement (Vsite). After inclusion of any correction factors relevant to the
measurement, the figure of loss which results from the verification procedure is known as "Site Attenuation".
NSA is determined from the value of Site Attenuation by subtraction of the antenna factors and mutual coupling effects.
The subtraction of the antenna factors and any mutual coupling effects makes NSA independent of antenna type.
Symbolically:
NSA = Vdirect - Vsite - AFT - AFR - AFTOT
where: Vdirect = received voltage using the "in-line" adapter;
Vsite = received voltage using the antennas;
AFT = antenna factor of the transmitting antenna;
AFR = antenna factor of the receiving antenna;
AFTOT = mutual coupling correction factor.
The verification procedure measures both Vdirect and Vsite and then (after relevant corrections and calculations) compares
the measured value of NSA against the theoretical figure calculated for that particular type of facility. The difference
between the two values at any specific frequency is a measure of the quality of the facility at that frequency.
6.8.2.1 Uncertainty contributions: Stage 1: Direct attenuation measurement

The first stage of the verification procedure is the direct attenuation measurement. This is carried out with all the items
of test equipment connected directly together via an "in line" adapter between the attenuators as shown in figure 21. The
components shown shaded are common to both stages of the procedure.
cable 1 cable 2
Signal Attenuator 1 "In line" Attenuator 2 Receiving
generator 10 dB adapter 10 dB device
ferrite beads ferrite beads
Figure 21: Stage 1: Direct attenuation measurement
Despite the commonality of most of the components to both stages of this procedure, the mismatch uncertainty
contribution for both stages has to be calculated and included in the uncertainty calculations. This is a result of load
conditions varying (i.e. antennas replacing the adapter in the second stage). Conversely, as a result of this commonality,
the uncertainty contributions of some of the individual components will cancel.
ETSI
74 ETSI TR 100 028-1 V1.4.1 (2001-12)
Whereas figure 21 shows, schematically, the test equipment set-up for this stage of the verification procedure, an
analysis diagram of the individual components (each of which contributes its own uncertainty) for this stage of the
measurement is shown in figure 22. Again, as stated above, the shaded areas represent components common to both
stages of the verification procedure.
6.8.2.1.1 Contributions from the mismatch components

Mismatch: direct attenuation measurement: The value of the combined standard uncertainty of the contribution due
to the mismatch from the source to the receptor, i.e. between the signal generator and the receiving device, is calculated
from the approach described in annex G. All the individual contributions are U-distributed.
NOTE 1: In this example the value taken for the signal generator output reflection coefficient is the worst case
value over the frequency band of interest. Similarly for the cable, adapter and attenuator VSWRs.
NOTE 2: The attenuation values of the cables and attenuators should be obtained from the data sheets/calibration
records at the specific frequency of the test, along with the associated uncertainties for these values.
Mismatch Mismatch
Attenuator 1 + Attenuator 2
+ Adaptor
10 dB 10 dB
Mismatch + + Mismatch
Cable Cable
+ +
factor factor
Transmitting Receiving
antenna Cable Cable antenna
cable 1 loss loss cable 2
source Receiving
device
Signal
generator device
Figure 22: Stage 1: Direct attenuation measurement individual uncertainty components
Signal generator: Output reflection coefficient: |ρG| = 0,20
Cable: Input and output reflection coefficients: |S11| = |S22| = 0,07
Attenuator: Input and output reflection coefficients |S11| = |S22| = 0,05
Attenuation = 10 dB |S12| = |S21| = 0,3162
Adapter: Input and output reflection coefficients |S11| = |S22| = 0,02
Attenuation = 0,1 dB |S12| = |S21| = 0,9886
ETSI
75 ETSI TR 100 028-1 V1.4.1 (2001-12)
Receiving device: Input reflection coefficient: |ρRD| = 0,20
Mismatch uncertainty in the direct attenuation measurement: In the following the transmitting antenna cable is
named cable 1, the transmitting antenna attenuator is named attenuator 1, the receiving antenna cable is named cable 2,
the receiving attenuator is named attenuator 2. Those components that are constant for both stages 1 and 2 are not
calculated as they do not contribute to the overall uncertainty.
Mismatch between:
- uj generator and cable 1: Constant for both stage 1 and 2. Hence this value does not contribute.
- uj cable 1 and attenuator 1: Constant for both stage 1 and 2. Hence this value does not contribute.
0,05 × 0,02 × 100

u j mismatch:attenuator1and adapter = % = 0,071 %
2
0,02 × 0,05 × 100

u j mismatch:adapter and attenuator1 = % = 0,071 %
2
- uj attenuator 2 and cable 2: Constant for both stage 1 and 2. Hence this value does not contribute.
- uj cable 2 and receiving device: Constant for both stage 1 and 2. Hence this value does not contribute.
- uj generator and attenuator 1: Constant for both stage 1 and 2. Hence this value does not contribute.
0,07 × 0,07 × 0,3162 × 100

u j mismatch:cable1and adapter = % = 0,035 %
2
0,05 × 0,05 × 0,9882 × 100

u j mismatch:attenuator1and attenuator2 = % = 0,173 %
2
0,02 × 0,07 × 0,3162 × 100

u j mismatch:adapter and cable2 = % = 0,010 %
2
- uj attenuator 2 and receiving device: Constant for both stage 1 and 2. Hence this value does not contribute.
0,2 × 0,02 × 0,8912 × 0,3162 × 100

u j mismatch:generatorand adapter = % = 0,022 %
2
0,07 × 0,05 × 0,3162 × 0,9882 × 100

u j mismatch:cable1and attenuator2 = % = 0,024 %
2
0,05 × 0,07 × 0,9882 × 0,3162 × 100

u j mismatch:attenuator1and cable2 = % = 0,024 %
2
0,02 × 0,2 × 0,3162 × 0,8912 × 100

u j mismatch:adapter and receivingdevice = % = 0,022 %
2
ETSI
76 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,2 × 0,05 × 0,8912 × 0,3162 × 0,9882 × 100

u j mismatch:generatorand attenuator2 = % = 0,055 %
2
- uj mismatch: cable 1 and cable 2: Less than 0,01 % due to the two attenuators, therefore neglected.
0,05 × 0,2 × 0,9882 × 0,3162 × 0,8912 × 100

u j mismatch:attenuator1and receivingdevice = % = 0,055 %
2
- uj mismatch: generator and cable 2: Less than 0,01 % due to the two attenuators, therefore neglected.
- uj mismatch: cable 1 and receiving device: Less than 0,01 % due to the two attenuators, therefore neglected.
- uj mismatch: generator and receiving device: Less than 0,01 % due to the two attenuators, therefore neglected.
The combined standard uncertainty of the mismatch is then calculated:
uc mismatch:direct att. = 0,0712 + 0,0712 + ... + 0,0552 + 0,0552 = 0,306 %
Transforming to logarithmic form (TR 100 028-2 [8], annex E): 0,306 %/11,5 = 0,026 dB.
The standard uncertainty of the contribution, due to the mismatch in the direct attenuation measurement, is designated
throughout all parts of TR 102 273 [3] as uj35. Its value in this example is 0,026 dB.
6.8.2.1.2 Contributions from individual components
6.8.2.1.2.1 Signal generator
Signal generator: absolute output level: In a verification procedure, the signal generator's absolute level uncertainty
contributes equally to both stages of the measurement. The standard uncertainty of the contribution due to the signal
generator absolute output level uncertainty is designated throughout all parts of TR 102 273 [3] as uj38.
NOTE 1: In this example case the standard uncertainty of the contribution due to the signal generator absolute
output level uncertainty is taken to be 0,00 dB since, once the level has been set in stage one of the
procedure, the level is not further adjusted. The uncertainty is therefore assumed to be systematic i.e. it
produces the same offset in both stages.
Signal generator: output level stability: In any test in which the contribution of the absolute level uncertainty of the
signal generator contributes to the combined standard uncertainty of the test i.e. it does not cancel due to the
methodology, the contribution from the output level stability is considered to have been included in the signal generator
absolute output level, uj38. Conversely, for any level in which the absolute level uncertainty of the signal generator does
not contribute to the combined standard uncertainty, the output level stability of the signal generator should be included.
The standard uncertainty of the contribution due to the signal generator output level stability is designated throughout
all parts of TR 102 273 [3] as uj39. Its value can be derived from manufacturers' data sheet.
NOTE 2: In this example case the uncertainty of the contribution due to the signal generator output level stability is
obtained from the manufacturers data sheet as ±0,02 dB. As nothing is said about the distribution of this
uncertainty, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is
assumed, and the standard uncertainty is calculated as 0,01155 dB. This is rounded down to 0,01 dB.
6.8.2.1.2.2 Transmitting antenna cable
Insertion loss: transmitting antenna cable: The transmitting antenna cable has an insertion loss as well as an
uncertainty associated with the measurement of its magnitude. The value of insertion loss and its uncertainty remain
valid provided the cable is not used outside the manufacturer's specification. At any given frequency the insertion loss
acts as a systematic offset and contributes equally to both stages of the measurement. The standard uncertainty of the
contribution due to the insertion loss uncertainty of the transmitting antenna cable is designated throughout all parts of
TR 102 273 [3] as uj41.
ETSI
77 ETSI TR 100 028-1 V1.4.1 (2001-12)
NOTE 1: In this example case the standard uncertainty of the contribution due to the insertion loss uncertainty of
the transmitting antenna cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
same offset in both stages.
Cable factor: transmitting antenna: Cable factor is defined as the total effect of the antenna cable's influence on the
measuring system including its interaction with the site. It consists of the leakage caused by cable screening
inefficiency, parasitic effects on the transmitting antenna (acting as a director or reflector) and introducing an
unbalanced, common mode current into the dipole balun. In a radiated measurement the standard uncertainty of the
cable factor, associated with each cable, is 0,5 dB provided that the precautions detailed in the methods have been
observed i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the standard uncertainty is
4,0 dB (justification for these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The standard uncertainty
of the contribution due to the cable factor of the transmitting antenna cable is designated throughout all parts of TR
102 273 [3] as uj19.
NOTE 2: In this example case the standard uncertainty of the contribution due to the cable factor of the transmitting
antenna cable is taken as 0,00 dB since there are no external fields involved other than leakage, which is
assumed to have a negligible effect on the measurement.
6.8.2.1.2.3 Transmitting antenna attenuator
Insertion loss: transmitting antenna attenuator: The transmitting antenna attenuator has an insertion loss as well as
an uncertainty associated with the measurement of its magnitude. The value of insertion loss and its uncertainty remain
valid provided the attenuator is not used outside the manufacturer's specification. At any given frequency the insertion
loss acts as a systematic offset and contributes equally to both stages of the measurement. The standard uncertainty of
the contribution, due to the insertion loss uncertainty of the transmitting antenna attenuator, is designated throughout all
parts of TR 102 273 [3] as uj40.
NOTE: In this example case the standard uncertainty of the contribution due to the insertion loss uncertainty of
the transmitting antenna attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces
the same offset in both stages.
6.8.2.1.2.4 Adapter
Insertion loss: adapter: The adapter has an insertion loss as well as an uncertainty associated with the measurement of
its magnitude. The value of insertion loss and its uncertainty remain valid provided the adapter is not used outside the
manufacturer's specification. The standard uncertainty of the contribution due to the insertion loss uncertainty of the
adapter is designated throughout all parts of TR 102 273 [3] as uj42. Its value can be derived from the manufacturer's
data sheet.
NOTE: In this example case the uncertainty of the contribution due to the insertion loss uncertainty of the adapter
is obtained from the manufacturer's data sheet as ±0,10 dB. As nothing is said about the distribution of
this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is
assumed, and the standard uncertainty is calculated as 0,06 dB.
6.8.2.1.2.5 Receiving antenna attenuator
Insertion loss: receiving antenna attenuator: The attenuator has an insertion loss as well as an uncertainty associated
with the measurement of its magnitude. The value of insertion loss and its uncertainty remain valid provided the
attenuator is not used outside the manufacturer's specification. At any given frequency the insertion loss acts as a
systematic offset and contributes equally to both stages of the measurement. The standard uncertainty of the
contribution due to the insertion loss uncertainty of the receiving antenna attenuator is designated throughout all parts of
TR 102 273 [3] as uj40.
the receiving antenna attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
ETSI
78 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.2.1.2.6 Receiving antenna cable
Insertion loss: receiving antenna cable: The receiving antenna cable has an insertion loss as well as an uncertainty
associated with the measurement of its magnitude. The value of insertion loss and its uncertainty remain valid provided
the cable is not used outside the manufacturer's specification. At any given frequency the insertion loss acts as a
contribution, due to the insertion loss uncertainty of the receiving antenna cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
the receiving antenna cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
Cable factor: receiving antenna cable: Cable factor is defined as the total effect of the antenna cable's influence on
the measuring system including its interaction with the site. It consists of the leakage caused by cable screening
inefficiency, parasitic effects on the receiving antenna (acting as a director or reflector) and introducing an unbalanced,
common mode current into the dipole balun. In a radiated measurement the standard uncertainty of the cable factor,
associated with each cable is 0,5 dB provided that the precautions detailed in the methods have been observed.
i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the standard uncertainty is 4,0 dB
(justification for these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The standard uncertainty of the
contribution due to the cable factor of receiving antenna cable is designated throughout all parts of TR 102 273 [3] as
uj19.
NOTE 2: In this example case the standard uncertainty of the contribution due to the cable factor of the receiving
antenna cable is taken as 0,00 dB since there are no external fields involved other than leakage, which is
6.8.2.1.2.7 Receiving device
In this, the first stage of the verification procedure, a reference level is recorded from the receiving device for a
particular output level from the signal generator. In the second stage (where the path loss between the two antennas is
measured), a second level is recorded on the receiving device. Only in the second stage do the linearity and absolute
level uncertainties of the receiver become involved in the calculation of the combined standard uncertainty of the
measurement.
Receiving device: absolute level: The standard uncertainty of the contribution due to the absolute level uncertainty of
the receiving device is designated throughout all parts of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution due to the absolute level uncertainty of
the receiving device is assumed to be 0,00 dB since, in this part of the measurement, a reference level is
recorded.
Receiving device: linearity: The standard uncertainty of the contribution due to the receiving device linearity is
designated throughout all parts of TR 102 273 [3] as uj48.
NOTE 2: In this example case the standard uncertainty of the contribution due to the receiving device linearity is
assumed to be 0,00 dB since, in this part of the measurement, a reference level is recorded.
6.8.2.1.3 Contribution from the random component

Random uncertainty: The magnitude can be assessed from multiple measurements of the direct attenuation
measurement. The standard uncertainty of the contribution due to the random uncertainty is designated throughout all
parts of TR 102 273 [3] as ui01. Its value can then be calculated.
The direct attenuation measurement was repeated 10 times. The following results were obtained in dBµV (before
correcting for cabling and attenuator network insertion loss):
- 106,8; 107,2; 106,7; 107,0; 107,2; 106,7; 107,1; 106,8; 107,1; 107,0.
Converting to linear terms:
- 0,2188; 0,2291; 0,2163; 0,2239; 0,2291; 0,2163; 0,2265; 0,2188; 0,2265; 0,2239.
ETSI
79 ETSI TR 100 028-1 V1.4.1 (2001-12)
- X = the sum of the measured values = 2,2292 V;
- Y = the sum of the squares of the measured values = 0,4972 V2.
X2 2,2292 2
Y− 0,4972 −
u c random = n = 10 = 5,444 ×10-3 (formula 5.6)
n −1 10 − 1
As the result is obtained as the mean value of 10 measurements and the standard uncertainty of the random uncertainty
is:
5,444 × 10 −3 100
u j random = × = 0,212 dB
0,22292 11,5
NOTE: In this example case the standard uncertainty of the contribution due to the random uncertainty is
0,212 dB. See also the note in clause 6.4.7.
6.8.2.1.4 Summary table of contributory components

The uncertainty contributions for stage 1 of the verification procedure are listed in table 3.
Table 3: Contributions from the direct attenuation measurement
uj or i Description of uncertainty contributions dB

uj35 mismatch: direct attenuation measurement 0,03
uj38 signal generator: absolute output level 0,00
uj39 signal generator: output level stability 0,01
uj41 insertion loss: transmitting antenna cable 0,00
uj19 cable factor: transmitting antenna 0,00
uj40 insertion loss: transmitting antenna attenuator 0,00
uj42 insertion loss: adapter 0,06
uj40 insertion loss: receiving antenna attenuator 0,00
uj41 insertion loss: receiving antenna cable 0,00
uj19 cable factor: receiving antenna 0,00
uj47 receiving device: absolute level 0,00
uj48 receiving device: linearity 0,00
ui01 random uncertainty (see note in clause 6.4.7) 0,21
The standard uncertainties from table 3 should be combined by RSS in accordance with TR 102 273 [3], part 1,
sub-part 1, clause 5. This gives the combined standard uncertainty (uc direct attenuation measurement) for the direct
attenuation measurement in dB.
The value of uc direct attenuation measurement is calculated as 0,221 dB.
6.8.2.2 Uncertainty contributions: Stage 2: Radiated attenuation measurement

The second stage of the verification procedure is the radiated attenuation measurement which is carried out by removing
the adapter and connecting each attenuator to an antenna as shown in figure 23, and recording the new level on the
receiving device. The difference in received levels (after allowance for any correction factors and calculations which
may be appropriate), for the same signal generator output level, reveals the NSA.
ETSI
80 ETSI TR 100 028-1 V1.4.1 (2001-12)
Transmitting antenna Receiving antenna
Signal cable 1 Attenuator 1 Attenuator 2 cable 2 Receiving

generator 10 dB 10 dB device
Ground plane
Figure 23: Stage 2: Radiated attenuation measurement
Whereas figure 23 shows, schematically, the test equipment set-up for this stage of the verification procedure an

Mismatch uncertainty transmitting and receiving parts: The value of the combined standard uncertainty of the
contribution due to the mismatch are calculated from the approach described in annex G. For this stage is calculated in
two parts. Firstly the standard uncertainty of the contribution due to the mismatch in the transmitting part, i.e. between
the signal generator, cable, attenuator and the transmitting antenna and secondly, that for the receiving part, i.e. between
the receiving antenna, attenuator, cable and the receiving device.
ETSI
81 ETSI TR 100 028-1 V1.4.1 (2001-12)
Transmitting antenna Receiving antenna

Direct
Antenna to antenna coupling path loss Antenna to antenna coupling
Reflected
path loss
Mutual image coupling Mutual image coupling
Antenna factor Site Antenna factor

Imperfections
Attenuator 2 (10 dB)

Attenuator 2 (10 dB)
Cable Cable
+ +
factor factor
antenna Cable loss Cable loss antenna
cable 1 cable 2
Mismatch +
+ Mismatch
source Signal generator Receiving device device
Figure 24: Stage 2: Radiated attenuation measurement individual uncertainty components
Mismatch: transmitting part:
Transmitting antenna: Input reflection coefficient: |ρTA| = 0,333
ETSI
82 ETSI TR 100 028-1 V1.4.1 (2001-12)
All these contributions are U-distributed. Those components that cancel are not calculated. Other contributions are (see
annex G):
- uj mismatch: generator and cable 1: Constant for both stage 1 and 2. Hence this value does not contribute.
- uj mismatch: cable 1 and attenuator 1: Constant for both stage 1 and 2. Hence this value does not contribute.
0,05 × 0,333 × 100

u j mismatch:attenuator1and antenna = % = 1,177 %
2
0,07 × 0,333 × 0,3162 × 100

u j mismatch:cable1and antenna = % = 0,165 %
2
0,2 × 0,333 × 0,8912 × 0,3162 × 100

u j mismatch:generatorand antenna = % = 0,373 %
2
uc mismatch:transmitting part = 1,1772 + 0,1652 + 0,3732 = 1,25%
transforming to the logarithmic form (TR 100 028-2 [8], annex E): 1,25 %/11,5 = 0,11 dB.
The standard uncertainty of the contribution due to the mismatch in the transmitting part, is designated throughout all
parts of TR 102 273 [3] as uj36. Its value in this example is 0,11 dB.
Mismatch: receiving part:
Receiving antenna: Input reflection coefficient: |ρRA| = 0,333
0,333 × 0,05 × 100

u j mismatch:antennaand attenuator2 = % = 1,177 %
2
0,333 × 0,07 × 0,316 2 × 100

u j mismatch:antenna and cable 2 = % = 0,165 %
2
0,333 × 0,2 × 0,316 2 × 0,8912 × 100

u j mismatch:antenna and receiving device = % = 0,373 %
2
ETSI
83 ETSI TR 100 028-1 V1.4.1 (2001-12)
u cmismatch:receiving part = 1,177 2 + 0,165 2 + 0,3732 = 1,25 %
Transforming to the logarithmic form (TR 100 028-2 [8], annex E): 1,25 %/11,5 = 0,11 dB.
The standard uncertainty of the contribution due to the mismatch in the receiving part is designated throughout all parts
of TR 102 273 [3] as uj37. Its value in this example is 0,11 dB.
6.8.2.2.2.2 Transmitting antenna cable
Insertion loss: transmitting antenna cable: The transmitting antenna cable has an insertion loss as well as an
contribution, due to the insertion loss uncertainty of the transmitting antenna cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
Cable factor: transmitting antenna cable: Cable factor is defined as the total effect of the antenna cable's influence
on the measuring system including its interaction with the site. It consists of the leakage caused by cable screening
inefficiency, parasitic effects on the transmitting antenna (acting as a director or reflector) and introducing an
unbalanced, common mode current into the dipole balun. In a radiated measurement the standard uncertainty of the
cable factor, associated with each cable, is 0,5 dB provided that the precautions detailed in the methods have been
observed. i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the standard uncertainty
is 4,0 dB (justification for these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The standard uncertainty
of the contribution due to the cable factor of the receiving antenna cable is designated throughout all parts of TR
102 273 [3] as uj19.
antenna cable is taken as 0,50 dB since the precautions detailed in the methods are assumed to have been
observed.
ETSI
84 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.2.2.2.3 Transmitting antenna attenuator
Insertion loss: transmitting antenna attenuator: The transmitting antenna attenuator has an insertion loss as well as
an uncertainty associated with the measurement of its magnitude. The value of insertion loss and its uncertainty remain
the contribution, due to the insertion loss uncertainty of the transmitting antenna attenuator, is designated throughout all
the transmitting antenna attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces
6.8.2.2.2.4 Transmitting antenna
Antenna: antenna factor of the transmitting antenna: Uncertainty is introduced as a result of the inaccurate
knowledge of the antenna factor of the transmitting antenna. The antenna factor contributes only to the radiated part of
this procedure. The standard uncertainty of the contribution due to the antenna factor uncertainty of the transmitting
antenna is designated throughout all parts of TR 102 273 [3] as uj44. For ANSI dipoles the value should be obtained
from table 4.
Table 4: Uncertainty contribution of the antenna factor of the transmitting antenna
Frequency Standard uncertainty of the

contribution
30 MHz ≤ frequency < 80 MHz 1,73 dB
frequency ≥ 180 MHz 0,30 dB
NOTE 1: For other antenna types the values should be taken from manufacturers' data sheets. If a value is not given
the standard uncertainty is 1,0 dB.
NOTE 2: In this example case the standard uncertainty of the contribution due to the antenna factor uncertainty of
the transmitting antenna is 0,30 dB since ANSI dipoles have been used and the frequency is above
180 MHz.
Antenna: tuning of the transmitting antenna: Uncertainty is introduced as a result of the inaccurate tuning of the
transmitting antenna. This only occurs in stage two of the measurement and therefore only contributes to this stage. The
standard uncertainty of the contribution, due to the tuning uncertainty of the transmitting antenna, is designated
throughout all parts of TR 102 273 [3] as uj46.
NOTE 3: In this example case the standard uncertainty of the contribution due to the tuning uncertainty of the
transmitting antenna is taken as 0,06 dB (see annex A).
Position of the phase centre: transmitting antenna: Uncertainty is introduced as a result of the inaccurate positioning
of the phase centre of the transmitting antenna. This only occurs in stage two of the measurement. The standard
uncertainty of the contribution, due to the uncertainty in the position of the phase centre of the transmitting antenna, is
NOTE 4: In this example case the standard uncertainty of the contribution due to the uncertainty in the position of
the phase centre of the transmitting antenna has been calculated from
(±(the offset from axis of rotation)/(range length) x 100 %). The positioning uncertainty is ±0,01 m and
therefore the worst case uncertainty = 0,01/3,0 = 0,333 %. As the offset can be anywhere between these
limits, the uncertainty is taken to be rectangularly distributed (see TR 102 273 [3], part 1, sub-part 1,
clause 5.1.2) and the standard uncertainty is calculated as 0,192 %. This is transformed to the logarithmic
form (TR 100 028-2 [8], annex E), to be 0,02 dB.
ETSI
85 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.2.2.2.5 Site factors
Ambient effect: Uncertainty is introduced as a result of local ambient signals raising the noise floor at the measurement
frequency. The standard uncertainty of the contribution due to the ambient effect is designated throughout all parts of
TR 102 273 [3] as uj34. The values of the standard uncertainty should be taken from table 5.
Table 5: Uncertainty contribution: Ambient effect
Receiving device noise floor Standard uncertainty of

(generator OFF) is within: the contribution
3 dB of measurement 1,57 dB
3 dB to 6 dB of measurement 0,80 dB
20 dB or more of the measurement 0,00 dB
NOTE 1: In this example case the standard uncertainty of the contribution due to the ambient effect is taken as
0,00 dB, since the chamber is assumed to be shielded.
Mutual coupling: transmitting antenna to its images in the absorbing material: This uncertainty is the effect of the
change produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution due to the
mutual coupling of the transmitting antenna to its images in the absorbing materials is designated throughout all parts of
TR 102 273 [3] as uj07.
NOTE 2: In this example case the standard uncertainty of the contribution due to the mutual coupling of the
transmitting antenna to its images in the absorbing materials is taken as 0,5 dB (see annex A).
Mutual coupling: transmitting antenna to its image in the ground plane: This uncertainty is the effect of the change
produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution due to the mutual
coupling of the transmitting antenna to its image in the ground plane, designated throughout all parts of TR 102 273 [3]
as uj15, has a value of 0,00 dB for ANSI dipoles since it is included, where significant, in the mutual coupling and
mismatch loss correction factors (see table A.20). For other dipoles the value can be obtained from table 6.
Table 6: Uncertainty contribution of the mutual coupling between

the transmitting antenna to its image in the ground plane
Spacing between the antenna Standard uncertainty

and the ground plane of the contribution
For a vertically polarized antenna
spacing ≤ 1,25 λ 0,15 dB
spacing > 1,25 λ 0,06 dB
For a horizontally polarized antenna
spacing < λ/2 1,15 dB
λ/2 ≤ spacing < 3λ/2 0,58 dB
3λ/2 ≤ spacing < 3λ 0,29 dB
spacing ≥ 3λ 0,15 dB
NOTE 3: In this example case the standard uncertainty of the contribution due to mutual coupling between the
transmitting antenna and its image in the ground plane is taken as 0,00 dB as we are assuming the use of
ANSI dipoles.
Mutual coupling: transmitting antenna to receiving antenna: This is the effect produced by any change in the gains
of the antennas which results from their close spacing. The standard uncertainty of the contribution due to the mutual
coupling of the transmitting antenna to receiving antenna, is designated throughout all parts of TR 102 273 [3] as uj10. It
has a standard uncertainty of 0,00 dB for ANSI dipoles since it is included, where significant, in the mutual coupling
and mismatch loss correction factors. For non-ANSI dipoles the standard uncertainty can be taken from table 7.
ETSI
86 ETSI TR 100 028-1 V1.4.1 (2001-12)

the transmitting and receiving antenna
Frequency Standard uncertainty Standard uncertainty

of the contribution of the contribution (10 m
(3 m range) range)
30 MHz ≤ frequency < 80 MHz 1,73 dB 0,60 dB
frequency ≥ 180 MHz 0,00 dB 0,00 dB
transmitting and receiving antennas is taken as 0,00 dB as we are using ANSI dipoles.
Mutual coupling: interpolation of mutual coupling and mismatch loss correction factors, only for ANSI dipoles:
The standard uncertainty of the contribution, due to the interpolation of mutual coupling and mismatch loss correction
factors, is designated throughout all parts of TR 102 273 [3] as uj12. It has, for spot frequencies given in table A.20, a
value of 0,00 dB. However, for all other frequencies, the standard uncertainty should be obtained from table 8.
Table 8: Uncertainty contribution of the interpolation of mutual coupling

and mismatch loss correction factors
Frequency (MHz) Standard uncertainty of

the contribution
for a spot frequency given in the table 0,00 dB
NOTE 5: In this example case the standard uncertainty of the contribution, due to the interpolation of mutual
coupling and mismatch loss correction factors, is taken as 0,00 dB as the frequency is above 180 MHz.
Range length: This contribution is associated with the curvature of the phase front from the transmitting antenna to the
receiving antenna. The standard uncertainty of the contribution, due to range length, is designated throughout all parts
of TR 102 273 [3] as uj16. The standard uncertainty is 0,00 dB if ANSI dipoles are used. For other types of antenna the
standard uncertainty of the contribution should be obtained from table 9.
Table 9: Uncertainty contribution of the range length (verification)

Range length (i.e. the horizontal distance Standard uncertainty of the
between phase centres) contribution
(d1 + d2) /4λ ≤ range length < (d1 + d2) /2λ
2 2
1,26 dB
(d1 + d2) /2λ ≤ range length < (d1 + d2) /λ
2 2
0,30 dB
(d1 + d2) /λ ≤ range length < 2(d1 + d2) /λ
2 2
0,10 dB
range length ≥ 2(d1 + d2) /λ
2
0,00 dB
NOTE 6: In table 9, d1 and d2 are the maximum dimensions of the antennas.
NOTE 7: In this example case the standard uncertainty of the contribution, due to the range length, is taken as
0,00 dB as we are using ANSI dipoles.
Reflectivity of absorbing material: transmitting antenna to the receiving antenna: This uncertainty is associated
with the magnitude of the reflections occurring from the side walls, end walls and ceiling. These magnitudes are a
function of the quality of the absorber at the frequency of test. The standard uncertainty of the contribution, due to
reflectivity of the absorber material between the transmitting antenna and the receiving antenna, is designated
throughout all parts of TR 102 273 [3] as uj03. The relevant value for this contribution should be taken from table 10.
ETSI
87 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 10: Uncertainty contribution of the reflectivity of absorbing material

between the transmitting and receiving antennas
Reflectivity of the absorbing material Standard uncertainty of the contribution

reflectivity < 10 dB 4,76 dB
10 dB ≤ reflectivity < 15 dB 3,92 dB
reflectivity ≥ 30 dB 0,74 dB
NOTE 8: In this example case the standard uncertainty of the contribution, due to the reflectivity of absorber
material between the transmitting antenna and the receiving antenna, is taken as 2,56 dB since it is
assumed that the absorber has a reflectivity in the range 15 dB to 20 dB.
Mutual coupling: receiving antenna to its images in the absorbing material: This uncertainty is the effect of the
change produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution, due to the
mutual coupling of the receiving antenna to its images in the absorbing material, is designated throughout all parts of
TR 102 273 [3] as uj07.
receiving antenna to its images in the absorbing material is taken as 0,5 dB.
Mutual coupling: receiving antenna to its image in the ground plane: This uncertainty is the effect of the change
produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution, due to the mutual
coupling of the receiving antenna to its image in the ground plane, is designated throughout all parts of TR 102 273 [3]
as uj15. It has a value of 0,00 dB for ANSI dipoles since it is included, where significant, in the mutual coupling and
mismatch loss correction factors. For other antennas the value can be obtained from table 11.

the receiving antenna and its image in the ground plane

λ/2 ≤ spacing < 3λ/2 0,58 dB
NOTE 10: In this example case the standard uncertainty of the contribution due to the mutual coupling between the
receiving dipole and its image in the ground plane is taken as 0,00 dB as we are using ANSI dipoles.
6.8.2.2.2.6 Receiving antenna
Correction: measurement distance: For verification procedures only one stage involves a radiated measurement and
hence no correction can be applied i.e. the uncertainty contribution is 0,00 dB. The standard uncertainty of the
contribution, due to the correction for measurement distance, is designated throughout all parts of TR 102 273 [3] as
uj18.
Correction: off boresight angle in elevation plane: For verification procedures only one stage involves a radiated
measurement and hence no correction can be applied i.e. the uncertainty contribution is 0,00 dB. The standard
uncertainty of the contribution, due to the correction for off boresight angle in elevation plane is designated throughout
all parts of TR 102 273 [3] as uj17.
Antenna: antenna factor of the receiving antenna: Uncertainty is introduced as a result of the inaccurate knowledge
of the antenna factor of the receiving antenna. The antenna factor contributes only to the radiated part of this procedure.
The standard uncertainty of the contribution due to the antenna factor uncertainty of the receiving antenna is designated
throughout all parts of TR 102 273 [3] as uj44. For ANSI dipoles the value should be obtained from table 12.
ETSI
88 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 12: Uncertainty contribution of the antenna factor of the receiving antenna
Frequency Standard uncertainty of the contribution

NOTE 1: For other antenna types the figures should be taken from manufacturers data sheets. If a figure is not
given the standard uncertainty is 1,0 dB.
the receiving antenna is 0,30 dB since ANSI dipoles have been used and the frequency is above
180 MHz.
Antenna: tuning of the receiving antenna: Uncertainty is introduced as a result of the inaccurate tuning of the
receiving antenna. This only occurs in stage two of the measurement and therefore does not cancel. The standard
uncertainty of the contribution, due to the tuning uncertainty of the receiving antenna, is designated throughout all parts
of TR 102 273 [3] as uj46.
NOTE 3: In this example case the standard uncertainty of the contribution due to the tuning uncertainty of the
receiving antenna is taken as 0,06 dB (see annex A).
Position of the phase centre: receiving antenna: Uncertainty is introduced as a result of the inaccurate positioning of
the phase centre of the receiving antenna. This only occurs in stage two of the measurement. The standard uncertainty
of the contribution, due to the uncertainty in the position of the phase centre of the receiving antenna, is designated
NOTE 4: In this example case the standard uncertainty of the contribution due to the position of the phase centre of
the receiving antenna has been calculated from ± (the offset) / (range length) x100 %). The positioning
uncertainty is ±0,01 m and therefore the worst case uncertainty = 0,01/3,0 = 0,333 %. As the offset can be
anywhere between these limits, the uncertainty is taken to be rectangularly distributed (see
TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) and the standard uncertainty is calculated as 0,192 %.
This is transformed to the logarithmic form (TR 100 028-2 [8], annex E), to be 0,02 dB.
6.8.2.2.2.7 Receiving antenna attenuator
Insertion loss: receiving antenna attenuator: The attenuator has an insertion loss as well as an uncertainty associated
contribution due to the insertion loss uncertainty of the receiving antenna attenuator is designated throughout all parts of
TR 102 273 [3] as uj40.
NOTE: In this example case the standard uncertainty of the contribution, due to the insertion loss uncertainty of
the transmitting antenna attenuator, is taken as 0,00 dB since the uncertainty is systematic i.e. it produces
6.8.2.2.2.8 Receiving antenna cable
Insertion loss: receiving antenna cable: The receiving antenna cable has an insertion loss as well as an uncertainty
contribution due to the insertion loss uncertainty of the receiving antenna cable is designated throughout all parts of
TR 102 273 [3] as uj41.
ETSI
89 ETSI TR 100 028-1 V1.4.1 (2001-12)
Cable factor: receiving antenna: Cable factor is defined as the total effect of the antenna cable's influence on the
inefficiency, parasitic effects on the receiving antenna (acting as a director or reflector) and introducing an unbalanced,
associated with each cable, is 0,5 dB provided that the precautions detailed in the methods have been observed.
contribution due to the cable factor of the receiving antenna cable is designated throughout all parts of TR 102 273 [3]
as uj19.
observed.
The first stage of the verification procedure involved setting a reference level on the receiving device for a particular
output level from the signal generator. In this the second stage (where the path loss between two antennas is measured),
a second level is obtained which results in linearity and absolute level uncertainties becoming involved in the
calculation of the combined standard uncertainty for the measurement.
Receiving device: absolute level: This uncertainty only contributes during the second stage of the procedure if the
input attenuation range setting on the receiving device has been changed from its setting in the first stage. The standard
uncertainty of the contribution due to the receiving device absolute level uncertainty is designated throughout all parts
of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution due to the receiving device absolute level
uncertainty (a range change is assumed) is obtained from the manufacturers data as ±1,0 dB. This is taken
as being rectangularly distributed (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2), so the standard
uncertainty is calculated as 0,58 dB.
Receiving device: linearity: The standard uncertainty of the contribution due to the receiving device linearity,
designated throughout all parts of TR 102 273 [3] as uj48, always contributes during the second stage of the procedure
unless there has been a range change in which case it is included in the receiving device absolute level uncertainty uj47.
NOTE 2: In this example case a range change has been assumed therefore the contribution is 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the radiated attenuation
measurements. The standard uncertainty of the contribution due to the random uncertainty is designated throughout all
parts of TR 102 273 [3] as ui01.
The radiated attenuation measurement was repeated 10 times. The following results were obtained in dBµV (before
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3.
- X = the sum of the measured values = 17,77 × 10-3;
- Y = the sum of the squares of the measured values = 32,10 × 10-6 V2.
ETSI
90 ETSI TR 100 028-1 V1.4.1 (2001-12)
Y−
X2
32,10 ×10 −6 −
17,77 × 10 −3 ( )2

n −1 10 − 1
is:
238,3 ×10 −6 100

−3
1,777 ×10 11,5
NOTE: In this example case the standard uncertainty of the contribution due to the random uncertainty is 1,17 dB.
See also the note in clause 6.4.7.

All the uncertainty contributions to this part of the procedure are listed in table 13.
Table 13: Contributions from the radiated attenuation measurement

uj36 mismatch: transmitting part 0,11
uj37 mismatch: receiving part 0,11
uj41 insertion loss: transmitting antenna cable 0,00
uj19 cable factor: transmitting antenna 0,50
uj40 insertion loss: transmitting antenna attenuator 0,00
uj44 antenna: antenna factor of the transmitting antenna 0,30
uj46 antenna: tuning of the transmitting antenna 0,06
uj22 position of the phase centre: transmitting antenna 0,02
uj34 ambient effect 0,00
uj07 mutual coupling: transmitting antenna to its images in the absorbing material 0,50
uj15 mutual coupling: transmitting antenna to its image in the ground plane 0,00
uj10 mutual coupling: transmitting antenna to the receiving antenna 0,00
uj12 mutual coupling: interpolation of mutual coupling and mismatch loss correction factors 0,00
uj16 range length 0,00
uj03 reflectivity of absorber material: transmitting antenna to the receiving antenna 2,56
uj07 mutual coupling: receiving antenna to its images in the absorbing material 0,50
uj15 mutual coupling: receiving antenna to its image in the ground plane 0,00
uj18 correction: measurement distance 0,00
uj17 correction: off boresight angle in the elevation plane 0,00
uj44 antenna: antenna factor of the receiving antenna 0,30
uj46 antenna: tuning of the receiving antenna 0,06
uj22 position of the phase centre: receiving antenna 0,02
uj40 insertion loss: receiving antenna attenuator 0,00
uj41 insertion loss: receiving antenna cable 0,00
uj19 cable factor: receiving antenna 0,50
sub-part 1, clause 5. This gives the combined standard uncertainty (uc NSA measurement) for the NSA measurement in dB.
The value of uc NSA measurement is calculated as 3,08 dB.
ETSI
91 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.2.2.5 Expanded uncertainty for the verification procedure

The combined standard uncertainty of the results of the verification procedure is the combination of the components
outlined in clauses 6.8.2.1.4 and 6.8.2.2.4. The components to be combined are uc direct attenuation measurement and
uc NSA measurement.
u c = 0,2212 + 3,08 2 = 3,08 dB
The expanded uncertainty is ±1,96 x 3,08 dB = ±6,04 dB at a 95 % confidence level.
6.8.3 Example 2: Measurement of a transmitter parameter (spurious

emission)
For the measurement of the effective radiated power in a spurious emission two stages of test are involved. The first
stage (the EUT measurement) is to measure on the receiving device, a level from the EUT. The second stage (the
substitution) involves replacing the EUT with a substitution antenna and signal source and adjusting the output level of
the signal generator until the same level as in stage one is achieved on the receiving device.
6.8.3.1 Uncertainty contributions: Stage 1: EUT measurement

The first stage of the spurious emission measurement is to measure on the receiving device, a level from the EUT. This
is normally carried out with a broadband antenna and receiver combination as shown in figure 25. The components
shown shaded are common to both stages of the test.
Test antenna
Test
antenna
cable 2
EUT Attenuator 2 Receiving
10 dB device
ferrite beads
Ground plane
Figure 25: Stage 1: EUT measurement
Due to the commonality of all of the components from the test antenna to the receiver in both stages of the test, the
mismatch uncertainty contributes identically to both stages and hence does not contribute to the combined standard
uncertainty of the measurement. Similarly, for the systematic uncertainty contributions (e.g. test antenna cable loss etc.)
of the individual components.
Whereas figure 25 shows, schematically, the test equipment set-up for this stage of the spurious emission measurement,
an analysis diagram of the individual components (each of which contributes its own uncertainty) for this stage of the
stages of the spurious emissions measurement.

Mismatch: receiving part: The uncertainty contribution due to the mismatch for the receiving part from the test
antenna to the receiver, can be calculated from the approach described in annex G. All the individual contributions are
U-distributed.
NOTE 1: In this example the value taken for the signal generator output reflection coefficient is the worst case
value over the frequency band of interest. Similarly for the cable, adapter and attenuator VSWRs.
ETSI
92 ETSI TR 100 028-1 V1.4.1 (2001-12)
The mismatch uncertainty between the test antenna and the receiving device however, contributes equally to both stages
of the test and therefore has no contribution to the combined standard uncertainty. Therefore it is not calculated.
Site factors Test antenna Mismatch
Antenna to Mutual
Antenna
EUT image +
gain
coupling coupling
Attenuator 2
10 dB
Direct Reflected
path loss path loss Mismatch +
Site Cable
effects factor +
Test
EUT
antenna Cable
loss
cable 2
Equipment
under test Mismatch +
Receiving
Receiving device
device
Figure 26: Stage 1: EUT measurement individual uncertainty components
The standard uncertainty of the contribution, due to the mismatch in the receiving part, is designated throughout all
NOTE 3: In this example case the standard uncertainty of the contribution due to mismatch in the receiving part is
taken as 0,00 dB, since the uncertainty is assumed to be systematic i.e. it is assumed constant and
common to both stages of the measurement.
6.8.3.1.2 Contributions from the individual components
6.8.3.1.2.1 EUT
EUT: influence of setting the power supply on the spurious emission level: This is the resulting uncertainty caused
by the uncertainty of setting of the power supply level. In this case normal supply conditions are assumed, not extreme.
The supply voltage uncertainty is taken to be ±100 mV so the uncertainty caused by this supply voltage uncertainty is
calculated using the dependency function (TR 100 028-2 [8], table C.1: "Equipment under test dependency functions
and uncertainties") whose mean value is 10 %/V and whose standard uncertainty is 3 %/V. The standard uncertainty of
the spurious emission level uncertainty caused by power supply voltage uncertainty (using formula 5.3) is:
( 0 ,1V )2
× (( 10%/V )2 +( 3%/V )2 ) = 0,60 %
3
This is then transformed to logarithmic form: 0,60 %/23,0 % = 0,03 dB (TR 100 028-2 [8], annex E).
ETSI
93 ETSI TR 100 028-1 V1.4.1 (2001-12)
The standard uncertainty of the contribution, due to the influence of setting the power supply on the spurious emission
level, is designated throughout all parts of TR 102 273 [3] as uj54.
NOTE 1: In this example case the standard uncertainty of the contribution, due to the influence of setting the power
supply on the spurious emission level, is calculated above as 0,03 dB.
EUT: influence of the ambient temperature on the spurious emissions: This is the uncertainty in the power level of
the spurious emission caused by the uncertainty in knowing the ambient temperature. The ambient temperature
uncertainty is ±1°C. The uncertainty caused by this temperature uncertainty is calculated using the dependency function
(TR 100 028-2 [8], table C.1: "EUT dependency functions and uncertainties") whose mean value is 4 %/°C and whose
standard uncertainty is 1,2 %/°C. The standard uncertainty of the spurious emission power level uncertainty caused by
ambient temperature uncertainty (using formula 5.3) is:
( 1°C)2
( ) × (( 4 ,0%/ °C )2 +( 1,2%/ °C )2 )= 2,41 %.
3
This is then transformed to logarithmic form: 2,41/23,0 % = 0,10 dB (TR 100 028-2 [8], annex E).
The standard uncertainty of the contribution, due to the influence of the ambient temperature on the spurious emissions,
is designated throughout all parts of TR 102 273 [3] as uj51.
NOTE 2: In this example case the standard uncertainty of the contribution due to the influence of the ambient
temperature on the spurious emissions level is calculated above as 0,10 dB.
EUT: mutual coupling to the power leads: This is the uncertainty associated with the influence (reflections, parasitic
effects, etc.) of the power leads on the EUT. The standard uncertainty associated with this effect is 0,5 dB provided that
the precautions detailed in the methods have been observed. I.e. routing and dressing of cables with ferrites. If no
prevention has been attempted the standard uncertainty is 2,0 dB. The standard uncertainty of the contribution, due to
the mutual coupling to the power leads, is designated throughout all parts of TR 102 273 [3] as uj54.
NOTE 3: In this example case the standard uncertainty of the contribution due to the mutual coupling to the power
leads is taken as 0,5 dB since the precautions detailed in the methods are assumed to have been observed.
Position of the phase centre: within the EUT volume: This contribution is associated with the uncertainty with which
the actual radiating point within the equipment volume is known. If this point is known exactly the contribution is
0,00 dB. The standard uncertainty of the contribution, due to the position of the phase centre within the EUT volume, is
NOTE 4: In this example case the standard uncertainty of the contribution due to the position of the phase centre
within the EUT volume has been calculated from (±(the maximum dimension of device)/(2 × range
length) × 100 %). In this example the position is not known. Hence, the uncertainty of the position of the
phase centre within the EUT of 0,15 m maximum dimension is 0,15/2 m = 0,075 m, and the worst case
uncertainty due to this offset is therefore ±(0,075/3,0) × 100 % = ±2,50 %. As the phase centre can be
anywhere inside the EUT, the uncertainty is taken as rectangularly distributed and the standard
uncertainty is calculated as 1,44 %. This is then transformed to the logarithmic form
(1,44/11,5) = 0,12 dB (TR 100 028-2 [8], annex E).
Positioning of the phase centre: within the EUT over the axis of rotation of the turntable: This contribution is
associated with the uncertainty with which the actual radiating point within the equipment is placed over the centre of
the turntable. If the point is placed exactly, the contribution is 0,00 dB. The standard uncertainty of the contribution due
to the positioning of the EUT phase centre over the axis of rotation of the turntable, is designated throughout all parts of
TR 102 273 [3] as uj21.
NOTE 5: In this example case the standard uncertainty of the contribution due to the positioning of the EUT phase
centre over the axis of rotation of the turntable is calculated from (± (the estimated offset from the axis of
rotation)/(2 × range length) × 100 %). In this case, the uncertainty of the positioning is taken as ±0,01 m,
and the worst case uncertainty is ±(0,01/3,0) × 100 = ±0,333 %. As the offset can be anywhere between
the limits the uncertainty is rectangularly distributed and the standard uncertainty is calculated as
0,192 %. This is then transformed to the logarithmic form (0,192/11,5) = 0,02 dB (TR 100 028-2 [8],
annex E).
ETSI
94 ETSI TR 100 028-1 V1.4.1 (2001-12)
Range length: This contribution is associated with the curvature of the phase front from the EUT to the test antenna.
The standard uncertainty of the contribution due to the range length is designated throughout all parts of
TR 102 273 [3] as uj16. The standard uncertainty of the contribution should be obtained from table 14.

2 2
1,26 dB
2 2
0,30 dB
2 2
0,10 dB
2
0,00 dB
NOTE 7: In this example case the standard uncertainty of the contribution due to the range length is taken as
0,00 dB since the range length is ≥ 2 (d1 + d2)2/λ.
Ambient effect: Ambient effect is the uncertainty caused by local ambient signals raising the noise floor at the
measurement frequency. The standard uncertainty of the contribution, due to the ambient effect, is designated
throughout all parts of TR 102 273 [3] as uj34. The values of the standard uncertainty should be taken from table 15.

(EUT OFF) is within: the contribution
Mutual coupling: amplitude effect of the test antenna on the EUT: This uncertainty results from the interaction
between the EUT and the test antenna when placed close together. The standard uncertainty of the contribution due to
the amplitude effect of the mutual coupling between the test antenna and the EUT, is designated throughout all parts of
TR 102 273 [3] as uj08. The standard uncertainty should be taken from table 16.
Table 16: Uncertainty contribution: Mutual coupling: amplitude effect

of the test antenna on the EUT
Range length Standard uncertainty of the contribution

0,62√((d1 + d2) /λ)≤ range length < 2(d1 + d2) /λ
3 2
0,50 dB
2
0,00 dB
NOTE 2: In this example case the standard uncertainty of the contribution due to the amplitude effect of the mutual
coupling between the test antenna and the EUT is 0,00 dB since the distance is ≥ 2 (d1 + d2)2/λ.
Mutual coupling: EUT to its images in the absorbing material: This uncertainty is dependant on the quality of the
absorbing material and the effect imaging of the EUT in the ceiling, side and end walls has on the input impedance
and/or gain of the integral antenna. The standard uncertainty of the contribution, due to the mutual coupling of the EUT
to its images in the absorbing material, is designated throughout all parts of TR 102 273 [3] as uj04.
NOTE 3: In this example case the standard uncertainty of the contribution due to the mutual coupling amplitude
effect of the absorbing material on the EUT is assumed to be 0,5 dB (see annex A).
ETSI
95 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mutual coupling: EUT to its image in the ground plane: This uncertainty results from the change in the EUT
spurious emission level as a result of being placed close to the ground plane. The standard uncertainty of the
contribution, due to the mutual coupling of the EUT to its image in the ground plane, is designated throughout all parts
of TR 102 273 [3] as uj13. Its value can be obtained from table 17.

the EUT to its image in the ground plane
Spacing between the EUT Standard uncertainty

For a vertically polarized EUT
For a horizontally polarized EUT
λ/2 ≤ spacing < 3λ/2 0,58 dB
NOTE 4: In this example case the standard uncertainty of the contribution, due to the mutual coupling of the EUT
to its image in the ground plane (assuming the polarization is vertical and the spacing above the ground
plane is < 1,25 λ at the test frequency), is taken as 0,15 dB.
Reflectivity of absorbing material: EUT to the test antenna. This uncertainty is associated with the magnitudes of
the reflections occurring from the side walls, end walls and ceiling. These magnitudes are a function of the quality of
the absorber at the frequency of test. The standard uncertainty of the contribution, due to the reflectivity of the
absorbing material between the EUT and the test antenna, is designated throughout all parts of TR 102 273 [3] as uj01.
The relevant value for this contribution should be taken from table 18.
Table 18: Uncertainty contribution of the reflectivity

of absorbing material between the EUT and test antenna
Reflectivity of the absorbing material Standard uncertainty of the contribution

reflectivity <10 dB 4,76 dB
10 ≤ reflectivity < 15 dB 3,92 dB
NOTE 5: In this example case the standard uncertainty of the contribution due to the reflectivity of the absorbing
material between the EUT and the test antenna is taken as 0,00 dB since this is a substitution
measurement and the contribution cancels (see annex A).
Mutual coupling: test antenna to its images in the absorbing material: This is the uncertainty due to the mutual
coupling between the test antenna and its images in the ceiling, side and end walls and is the effect of the change
produced in the antenna's input impedance and/or gain. As this is the first stage of a substitution measurement and the
uncertainty is common to both stages it will only contribute in the second stage if the test antenna is located at a
different height on the antenna mast. The standard uncertainty of the contribution, due to the mutual coupling between
the test antenna and its images in the absorbing material, is designated throughout all parts of TR 102 273 [3] as uj06.
test antenna and its images in the absorbing material is taken as 0,00 dB since this is the reference
position.
Mutual coupling: test antenna to its image in the ground plane: This is the uncertainty due to the mutual coupling
between the test antenna and its image in the ground plane and is the effect of the change produced in the antenna's
input impedance and/or gain when placed close to a ground plane. As this is the first stage of a substitution
measurement and the uncertainty is common to both stages, it will only contribute in the second stage if the test antenna
is located at a different height on the antenna mast. The standard uncertainty of the contribution, due to the mutual
coupling between the test antenna and its image in the ground plane, is designated throughout all parts of
TR 102 273 [3] as uj14.
ETSI
96 ETSI TR 100 028-1 V1.4.1 (2001-12)
test antenna and its image in the ground plane is taken as 0,00 dB since this is the reference position.
6.8.3.1.2.3 Test antenna
Correction: measurement distance: For those tests in which the test antenna on the mast peaks at different heights in
the two stages, a correction for the measurement distance should be made to account for the different measurement
distances. The standard uncertainty of the contribution due to the correction for measurement distance is designated
NOTE 1: In this example case the standard uncertainty of the contribution due to the correction for measurement
distance is taken as 0,00 dB since, in this, the first stage of the measurement, only a reference height is
being set.
Correction: off boresight angle in elevation plane: For those tests in which the test antenna on the mast peaks at
different heights in the two stages, a correction must be made to account for the different angles subtended by the
EUT/substitution antenna. The standard uncertainty of the contribution, due to the correction for off boresight angle in
elevation plane, is designated throughout all parts of TR 102 273 [3] as uj17.
NOTE 2: In this example case the standard uncertainty of the contribution due to the correction for off boresight
angle in elevation plane is taken as 0,00 dB since in this, the first stage of the measurement, only a
reference height is being set.
Antenna: gain of the test antenna: The gain, and its uncertainty, of the test antenna act as systematic offsets since they
are present in both stages of the test. The standard uncertainty of the contribution, due to the uncertainty of the gain of
the test antenna, is designated throughout all parts of TR 102 273 [3] as uj45.
NOTE 3: In this example case the standard uncertainty of the contribution due to the uncertainty of the gain of the
test antenna is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same offset in both
stages.
Antenna: tuning of the test antenna: This uncertainty is introduced as a result of inaccurate tuning of the test antenna.
The standard uncertainty of the contribution, due to the tuning of the test antenna, is designated throughout all parts of
TR 102 273 [3] as uj46.
NOTE 4: In this example case the standard uncertainty of the contribution due to the tuning of the test antenna is
taken as 0,00 dB since the uncertainty is systematic i.e. it is assumed constant and common to both stages
of the measurement and, provided that once set in stage one of the test it is not subsequently re-tuned, its
contribution is the same in both stages.
Position of the phase centre: test antenna: The horizontal position of the test antenna defines one end of the range
length. The standard uncertainty of the contribution, due to the position of the phase centre of the test antenna, is
NOTE 5: In this example case the test antenna is assumed to describe a vertical straight line as its height on the
mast is changed. The standard uncertainty of the contribution due to the position of the phase centre of the
test antenna is assumed to be 0,00 dB (see annex A).
6.8.3.1.2.4 Test antenna attenuator
Insertion loss: test antenna attenuator: The attenuator has an insertion loss as well as an uncertainty associated with
the measurement of its magnitude. The value of insertion loss and its uncertainty remain valid provided the attenuator is
not used outside the manufacturer's specification. At any given frequency the insertion loss acts as a systematic offset
and contributes equally to both stages of the measurement. The standard uncertainty of the contribution due to the
insertion loss of the test antenna attenuator is designated throughout all parts of TR 102 273 [3] as uj40.
the test antenna attenuator, is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same
offset in both stages.
ETSI
97 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.3.1.2.5 Test antenna cable
Insertion loss: test antenna cable: The test antenna cable has an insertion loss as well as an uncertainty associated
with the measurement of its magnitude. The value of insertion loss and its uncertainty remain valid provided the cable is
and contributes equally to both stages of the measurement. The standard uncertainty of the contribution, due to the
insertion loss uncertainty of the test antenna cable, is designated throughout all parts of TR 102 273 [3] as uj41.
the test antenna cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same
Cable factor: test antenna cable: Cable factor is defined as the total effect of the antenna cable's influence on the
inefficiency, parasitic effects (acting as a director or reflector) on the test antenna and introducing an unbalanced,
contribution due to the cable factor of the test antenna cable is designated throughout all parts of TR 102 273 [3] as uj19.
NOTE 2: In this example case the standard uncertainty of the contribution due to the cable factor of the test antenna
cable is taken as 0,50 dB since in this measurement, the cable changes position in both stages (the height
of the test antenna being optimized in both stages) even though the precautions detailed in the methods
have been observed.
Whereas the first stage of a spurious emission test is to observe and record the received level from the EUT on the
receiving device, the second stage involves the adjustment of the output level of a signal generator to achieve the same
received level from a substitution antenna. As a consequence of this methodology, the receiving device contributes
neither linearity nor absolute level uncertainty to either stage of the test.
Receiving device: absolute level: The standard uncertainty of the contribution, due to the absolute level uncertainty of
the receiving device, is designated throughout all parts of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution, due to the absolute level uncertainty of
the receiving device is taken as 0,00 dB.
Receiving device: linearity: The standard uncertainty of the contribution, due to the receiving device linearity, is
taken as 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the spurious emission. The
standard uncertainty of the contribution due to the random uncertainty is designated throughout all parts of
TR 102 273 [3] as ui01.
The EUT measurement was repeated 10 times. The following results were obtained in dBµV (before correcting for
cabling and attenuator network insertion loss):
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3.
ETSI
98 ETSI TR 100 028-1 V1.4.1 (2001-12)
Y−
X2
32,10 × 10 −6
−
(17,77 ×10 )−3 2
uc random = n = 10 = 238,3 ×10-6 (formula 5.6)

n −1 10 − 1
is:
238,3 ×10 −6 100

−3
1,777 ×10 11,5

All the uncertainty contributions for this part of the procedure are listed in table 19.
Table 19: Contributions from the EUT measurement

uj54 EUT: influence of setting the power supply on the spurious emission level 0,03
uj51 EUT: influence of the ambient temperature on the spurious emission level 0,03
uj55 EUT: mutual coupling to the power leads 0,50
uj20 position of the phase centre: within the EUT volume 0,12
uj21 positioning of the phase centre: within the EUT over the axis of rotation of the turntable 0,02
uj08 mutual coupling: amplitude effect of the test antenna on the EUT 0,00
uj04 Mutual coupling: EUT to its images in the absorbing material 0,50
uj13 mutual coupling: EUT to its image in the ground plane 0,15
uj01 reflectivity of absorbing material: EUT to the test antenna 0,00
uj06 mutual coupling: test antenna to its images in the absorbing material 0,00
uj14 mutual coupling: test antenna to its image in the ground plane 0,00
uj17 correction: off boresight angle in elevation plane 0,00
uj45 antenna: gain of the test antenna 0,00
uj46 antenna: tuning of the test antenna 0,00
uj22 position of the phase centre: test antenna 0,00
uj40 insertion loss: test antenna attenuator 0,00
uj41 insertion loss: test antenna cable 0,00
uj19 cable factor: test antenna cable 0,50
sub-part 1, clause 5. This gives the combined standard uncertainty (uc contribution from the EUT measurement) for the EUT
measurement in dB.
The value of uc contribution from the EUT measurement is calculated as 1,47 dB.
ETSI
99 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.3.2 Uncertainty contributions: Stage 2: Substitution measurement

The second stage of the spurious emission test (the substitution) involves replacing the EUT with a substitution antenna
and signal source as shown in figure 27, and adjusting the output level of the signal generator until the same level as in
stage one is achieved on the receiving device.
Test antenna
Test
antenna
Ground plane
Figure 27: Stage two: Typical emission substitution test
Whereas figure 27 shows, schematically, the test equipment set-up for this substitution stage of the spurious emission
test, figure 28, an analysis diagram, provides a detailed picture of the individual uncertainty components (each of which
contributes its own uncertainty) for this stage in the measurement. As stated above, the shaded areas represent
components common to both stages of the test method.

Mismatch uncertainty transmitting and receiving parts: The value of the combined standard uncertainty of the
contribution due to the mismatch for the substitution measurement are calculated from the approach described in
annex G. For this stage it is calculated in two parts. Firstly the standard uncertainty of the contribution due to the
mismatch in the transmitting part, i.e. between the signal generator, cable, attenuator and the substitution antenna and
secondly, that for the receiving part, i.e. between the test antenna, attenuator, cable and the receiver. However, only the
contribution for the transmitting part is actually calculated since the receiving part is common to both stages of the test
and its uncertainty contribution therefore largely cancels.
Transmitting antenna: Input reflection coefficient: |ρTA| = 0,333
annex G):
0,05 × 0,333 × 100

u j attenuator1and antenna = % = 1,177 %
2
ETSI
100 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,07 × 0,333 × 0,3162 × 100

u j cable1and antenna = % = 0,165 %
2
0,2 × 0,333 × 0,8912 × 0,3162 × 100

u j generatorand antenna = % = 0,373%
2
Substitution antenna Test antenna

Direct
Antenna to antenna coupling path loss Antenna to antenna coupling
Reflected
path loss

Imperfections
Mismatch + Mismatch
Attenuator 1 Attenuator 2
10 dB 10 dB
Mismatch Mismatch
+
Cable Cable
+
factor +
factor
Transmitting
Test
antenna Cable loss Cable loss
cable 1 antenna
cable 2
Mismatch
+ Mismatch
source Signal generator Receiving device device
Figure 28: Stage two: Substitution measurement individual uncertainty components
uc mismatch:substitution = 1,1772 + 0,1652 + 0,3732 = 1,25%
ETSI
101 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mismatch: Receiving part:

The mismatch uncertainty between the test antenna and the receiving device contributes equally to both stages of the
test and therefore has no contribution to the combined standard uncertainty. Therefore it is not calculated. The standard
uncertainty of the contribution due to mismatch in the receiving part is designated throughout all parts of
TR 102 273 [3] as uj37.
NOTE: In this example case the standard uncertainty of the contribution due to mismatch in the receiving part is
taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same offset in both stages.
Signal generator: absolute output level: The signal generator replaces the EUT in the substitution part of this test and,
as a result, should be included in the combined standard uncertainty since it does not cancel as a systematic offset. The
standard uncertainty of the contribution, due to the signal generator absolute output level, is designated throughout all
NOTE 1: In this example case the uncertainty of the contribution due to the signal generator absolute output level
uncertainty is obtained from the manufacturers data sheet as ±1,0 dB. As nothing is said about the
distribution of this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1,
clause 5.1.2) in logs is assumed, and the standard uncertainty is calculated as 0,58 dB.
NOTE 2: In this example case the standard uncertainty of the contribution due to the signal generator output level
stability is taken as 0,00 dB as it is covered by the absolute level uncertainty.
6.8.3.2.2.2 Substitution antenna cable
Insertion loss: substitution antenna cable: The substitution antenna cable has an insertion loss as well as an
contribution due to the insertion loss uncertainty of the substitution antenna cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
NOTE 1: In this example case the uncertainty of the contribution due to the insertion loss uncertainty of the
substitution antenna cable is taken from the manufacturers data sheet as ±0,5 dB. As nothing is said about
the distribution, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is
assumed and the standard uncertainty is calculated as 0,29 dB.
Cable factor: substitution antenna cable: Cable factor is defined as the total effect of the antenna cable's influence on
inefficiency, parasitic effects (acting as a director or reflector) and introducing an unbalanced, common mode current
into the dipole balun. In a radiated measurement the standard uncertainty of the cable factor, associated with the
substitution antenna cable is 0,5 dB provided the precautions detailed in the methods have been observed. i.e. routing
and dressing of cables with ferrites. If no prevention has been attempted the standard uncertainty is 4,0 dB (justification
for these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The standard uncertainty of the contribution,
due to the cable factor of the substitution antenna cable, is designated throughout all parts of TR 102 273 [3] as uj19.
NOTE 2: In this example case the standard uncertainty of the contribution due to the cable factor of the substitution
observed.
ETSI
102 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.3.2.2.3 Substitution antenna attenuator
Insertion loss: substitution antenna attenuator: The attenuator has an insertion loss as well as an uncertainty
the attenuator is not used outside the manufacturer's specification. At any given frequency the insertion loss is taken
from the manufacturer's data sheet since it does not cancel as a systematic offset (it only appears in one stage of the
test). The standard uncertainty of the contribution, due to the insertion loss uncertainty of the substitution antenna
attenuator, is designated throughout all parts of TR 102 273 [3] as uj40.
NOTE: In this example case the uncertainty of the contribution, due to the insertion loss uncertainty of the
substitution antenna attenuator, is obtained from the manufacturer's data sheet as ±0,3 dB. As nothing is
said about the distribution, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2)
in logs is assumed and the standard uncertainty is calculated as 0,17 dB.
6.8.3.2.2.4 Substitution antenna
Antenna: gain of the substitution antenna: The gain (and its uncertainty) of the substitution antenna is only involved
in the second stage of the test. The standard uncertainty of the contribution due to the gain of the substitution antenna is
designated throughout all parts of TR 102 273 [3] as uj45. For ANSI dipoles the value should be obtained from table 20.
Table 20: Uncertainty contribution: Antenna: gain of the test or substitution antenna

contribution
30 MHz ≤ frequency ≤ 80 MHz 1,73 dB
80 MHz < frequency ≤ 180 MHz 0,60 dB
frequency > 180 MHz 0,30 dB
NOTE 2: In this example case the standard uncertainty of the contribution due to the gain of the substitution
antenna is taken as 0,3 dB as an ANSI dipole is used and the frequency is above 180 MHz.
Antenna: tuning of the substitution antenna: Uncertainty is introduced as a result of the inaccurate tuning of the
substitution antenna. This only occurs in stage two of the measurement. The standard uncertainty of the contribution,
due to the tuning of the substitution antenna, is designated throughout all parts of TR 102 273 [3] as uj46.
NOTE 3: In this example case the standard uncertainty of the contribution due to the tuning of the substitution
antenna is taken as 0,06 dB.
Position of the phase centre: substitution antenna: Uncertainty is introduced as a result of the inaccurate positioning
of the phase centre of the substitution antenna. This only occurs in stage two of the measurement. The standard
uncertainty of the contribution, due to the position of the phase centre of the substitution antenna, is designated
NOTE 4: In this example case the standard uncertainty of the contribution due to the uncertainty in the position of
the phase centre of the substitution antenna has been calculated from
(±(the offset from axis of rotation)/ (range length) x100 %). The positioning uncertainty is ±0,01 m and
therefore the worst case uncertainty = 0,01/3,0 = 0,333 %. As the offset can be anywhere between these
limits, the uncertainty is taken to be rectangularly distributed (see TR 102 273 [3], part 1, sub-part 1,
clause 5.1.2) and the standard uncertainty is calculated as 0,192 %. This is transformed to the logarithmic
form (TR 100 028-2 [8], annex E), to be 0,02 dB.
ETSI
103 ETSI TR 100 028-1 V1.4.1 (2001-12)

Mutual coupling: substitution antenna to its images in the absorbing material: This uncertainty is the effect of the
change produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution due to the
mutual coupling of the substitution antenna to its images in the absorbing material is designated throughout all parts of
TR 102 273 [3] as uj06.
transmitting antenna to its images in the absorbing materials is taken as 0,5 dB (see annex A).
Mutual coupling: substitution antenna to its image in the ground plane: This uncertainty is the effect of the change
coupling of the substitution antenna to its image in the ground plane, designated throughout all parts of TR 102 273 [3]
as uj14. has a value of 0,00 dB for ANSI dipoles since it is included, where significant, in the mutual coupling and
mismatch loss correction factors (see table A.20). For other dipoles the value can be obtained from table 22.

the substitution antenna and its image in the ground plane

λ/2 ≤ spacing < 3λ/2 0,58 dB
substitution antenna and its image in the ground plane is taken as 0,58 dB.
Mutual coupling: substitution antenna to the test antenna: This is the effect produced by any change in gain of the
antennas which results from their close spacing. The standard uncertainty of the contribution due to the mutual coupling
of the substitution antenna to the test antenna, designated throughout all parts of TR 102 273 [3] as uj11. For ANSI
dipoles the value of this uncertainty is 0,00 dB as it is included, where significant, in the mutual coupling and mismatch
loss correction factors. For non-ANSI dipoles the standard uncertainty for frequencies can be taken from table 23.
ETSI
104 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 23: Uncertainty contribution of the mutual coupling

between the substitution and test antennas

(3 m range) range)
substitution and test antennas is taken as 0,00 dB as the frequency is above 180 MHz.
The standard uncertainty of the contribution due to the interpolation of mutual coupling and mismatch loss correction
factors, is designated throughout all parts of TR 102 273 [3] as uj12. For spot frequencies given in table A.20, the value
of the contribution is 0,00 dB. However, for all other frequencies, the standard uncertainty should be obtained from
table 24.
Table 24: Uncertainty contribution of the interpolation

of mutual coupling and mismatch loss correction factors
Frequency (MHz) Standard uncertainty of the contribution

NOTE 5: In this example case the standard uncertainty of the contribution due to the interpolation of mutual
coupling and mismatch loss correction factors is taken as 0,00 dB as the frequency is above 180 MHz.
Range length: This contribution is associated with the curvature of the phase front from the substitution antenna to the
test antenna. The standard uncertainty of the contribution, due to range length, is designated throughout all parts of
TR 102 273 [3] as uj16. The standard uncertainty is 0,00 dB if ANSI dipoles are used. For other types of antenna the

2 2
1,26 dB
2 2
0,30 dB
2 2
0,10 dB
2
0,00 dB
NOTE 7: The standard uncertainty of the contribution due to the range length is taken as 0,00 dB since in this case
the range length is ≥ 2 (d1 + d2)2/λ.
Reflectivity of absorbing material: substitution antenna to the test antenna: This uncertainty is associated with the
magnitude of the reflections occurring from the side walls, end walls and ceiling. These magnitudes are a function of
the quality of the absorber at the frequency of test. However, in this, a substitution measurement, the contribution is
only concerned with taking into account the possible differences in the antenna patterns (principally in the vertical
plane) between the EUT and substitution antenna. The standard uncertainty of the contribution due to the reflectivity of
the absorbing material between the substitution and test antenna is designated throughout all parts of TR 102 273 [3] as
uj02.
material between the substitution and test antenna is 0,50 dB (see annex A).
ETSI
105 ETSI TR 100 028-1 V1.4.1 (2001-12)
produced in the antenna's input impedance and/or gain. As this is the second stage of a substitution measurement and
the uncertainty is common to both stages it will only contribute in this stage if the test antenna is located at a different
height on the antenna mast from the first stage. The standard uncertainty of the contribution, due to the mutual coupling
of the test antenna to its images in the absorbing material, is designated throughout all parts of TR 102 273 [3] as uj06.
NOTE 9: In this example case the standard uncertainty of the contribution due to the mutual coupling of the test
antenna to its images in the absorbing material is taken as 0,50 dB since this is the second stage of a
substitution measurement and the test antenna is assumed to be located at a different height to stage one
and therefore the contribution does not cancel.
Mutual coupling: test antenna to its image in the ground plane: As this is the second stage of a substitution
measurement and the uncertainty is common to both stages it will only contribute in this stage if the test antenna is
located at a different height on the antenna mast from the first stage. The standard uncertainty of the contribution, due to
the mutual coupling of the test antenna to its image in the ground plane, is designated throughout all parts of
TR 102 273 [3] as uj14.
antenna to its image in the ground plane is taken as 0,50 dB since this is the second stage of a substitution
measurement and the test antenna is assumed to be located at a different height to stage one and therefore
the contribution does not cancel.
the two stages, a correction should be made to account for the different measurement distances. Where a correction is
required the standard uncertainty of the correction factor should be taken as 0,10 dB. The standard uncertainty of the
contribution due to the correction for measurement distance is designated throughout all parts of TR 102 273 [3] as uj18.
distance is taken as 0,10 dB since it is assumed that, in this second stage, a different height of test antenna
has been necessary and the correction applied.
different heights in the two stages, a correction should be made to account for the different angles subtended by the
EUT/substitution antenna. Where a correction is required the standard uncertainty of the correction factor should be
taken as 0,10 dB. The standard uncertainty of the contribution due to the correction for off boresight angle in elevation
plane is designated throughout all parts of TR 102 273 [3] as uj17.
angle in the elevation plane is taken as 0,10 dB since it is assumed that, in this second stage, a different
height of test antenna has been necessary and the correction applied.
stages.
TR 102 273 [3] as uj46.
ETSI
106 ETSI TR 100 028-1 V1.4.1 (2001-12)
and contributes equally to both stages of the measurement. The standard uncertainty of the contribution, due to the
cable is taken as 0,50 dB since in this measurement, the cable changes position in both stages (the height
of the test antenna being optimized in both stages) even though the precautions detailed in the methods
have been observed.
Whereas the first stage of a spurious emission test is to observe and record the received level from the EUT on the
receiving device, the second stage involves the adjustment of the output level of a signal generator to achieve the same
received level from a substitution antenna. As a consequence of this methodology, the receiving device contributes
neither linearity nor absolute level uncertainty to either stage of the test.
Receiving device: absolute level: The standard uncertainty of the contribution, due to the absolute level uncertainty of
the receiving device, is designated throughout all parts of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution, due to the absolute level uncertainty of
the receiving device is taken as 0,00 dB.
ETSI
107 ETSI TR 100 028-1 V1.4.1 (2001-12)
Receiving device: linearity: The standard uncertainty of the contribution, due to the receiving device linearity, is
taken as 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the substitution measurement.
The standard uncertainty of the contribution due to the random uncertainty is designated throughout all parts of
TR 102 273 [3] as ui01.
The substitution measurement was repeated 10 times. The following levels were set on the signal generator in dBm
(before correcting for cabling and attenuator network insertion loss):
- -20,1; -20,1; -20,2; -20,2; -20,1; -20,1; -20,2; -20,3; -20,3; -20,3.
- 9,772 × 10-3; 9,772 × 10-3; 9,550 × 10-3; 9,550 × 10-3; 9,772 × 10-3; 9,772 × 10-3; 9,550 × 10-3; 9,333 × 10-3;
9,333 × 10-3; 9,333 × 10-3.
- Y = the sum of the squares of the measured values = 916,89 × 10-6 W2.
Y−
X2
916,89 × 10 −6 −
(95,737 ×10 )
−3 2
u c random = n = 10 = 192,3 × 10-6 (formula 5.7)

n −1 10 − 1
is:
192,3 × 10 −6 100
−3
9,5737 × 10 23,0
ETSI
108 ETSI TR 100 028-1 V1.4.1 (2001-12)

All the uncertainties contributions for this part of the procedure are listed in table 26.
Table 26: Contributions from the substitution

uj41 insertion loss: substitution antenna cable 0,29
uj19 cable factor: substitution antenna cable 0,50
uj40 insertion loss: substitution antenna attenuator 0,17
uj45 antenna: gain of the substitution antenna 0,30
uj46 antenna: tuning of the substitution antenna 0,06
uj22 position of the phase centre: substitution antenna 0,02
uj06 mutual coupling: substitution antenna to its images in the absorbing material 0,50
uj14 mutual coupling: substitution antenna to its image in the ground plane 0,58
uj11 mutual coupling: substitution antenna to the test antenna 0,00
uj02 reflectivity of absorbing material: substitution antenna to the test antenna 0,50
sub-part 1, clause 5. This gives the combined standard uncertainty (uc substitution measurement) for the NSA measurement
in dB.
The value of uc substitution measurement is calculated as 1,56 dB.
6.8.3.2.5 Expanded uncertainty for the spurious emission test

The combined standard uncertainty of the results of the spurious emissions test is the combination of the components
outlined in clauses 6.8.3.1.4 and 6.8.3.2.4. The components to be combined are (uc EUT measurement) and (uc substitution
measurement).
u c = 1,47 2 + 1,56 2 = 2,15 dB
ETSI
109 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.4 Example 3: Measurement of a receiver parameter (Sensitivity)

For the measurement of receiver sensitivity two stages of test are involved. The first stage (determining the Transform
Factor of the site) involves measuring the field strength at the point where the receiver will be placed and determining
the relationship between the signal generator output power level and the resulting field strength. The second stage (the
EUT measurement) involves replacing the measuring antenna with the EUT and adjusting the output level of the signal
generator until the required response is obtained on the receiver. The signal generator output power level is then
converted to field strength using the Transform Factor.
6.8.4.1 Uncertainty contributions: Stage 1: Transform Factor measurement

The first stage of the receiver sensitivity test is to determine the Transform Factor of the site. This is normally carried
out by placing a measuring antenna in the volume occupied by the EUT and determining the relationship between the
signal generator output power and the resulting field strength. The test equipment configuration is shown in figure 29.
The components shown shaded are common to both stages of the test.
Measuring
antenna Test
cable 1 antenna
cable 2
Receiving Attenuator 1 Attenuator 2 Signal

device 10 dB 10 dB generator
Measuring Test antenna

antenna
Ground plane
Figure 29: Stage 1: Transform Factor measurement
mismatch uncertainty contributes identically to both stages and hence does not contribute to the combined standard
uncertainty of the measurement. Similarly, for the systematic uncertainty contributions (e.g. test antenna cable loss etc.)
of the individual components.
Whereas figure 29 shows, schematically, the equipment set-up for this stage of the receiver sensitivity test, an analysis
diagram of the individual components (each of which contributes its own uncertainty) for this stage of the measurement
is shown in figure 30. Again, as stated above, the shaded areas represent components common to both stages of the
receiver sensitivity test.

Mismatch in the transmitting and receiving part: The value of the combined standard uncertainty of the contribution
due to the mismatch are calculated from the approach described in annex G. For this stage is calculated in two parts.
Firstly the standard uncertainty of the contribution due to the mismatch in the transmitting part, i.e. between the signal
generator, cable, attenuator and the test antenna and secondly, that for the receiving part, i.e. between the receiving
antenna, attenuator, cable and the receiving device.
The standard uncertainty of the contribution due to mismatch in the transmitting part is designated throughout all parts
of TR 102 273 [3] as uj36.
NOTE: In this example case the uncertainty due to mismatch in the receiving part is taken as 0,00 dB since the
uncertainty is systematic i.e. it produces the same offset in both stages.
ETSI
110 ETSI TR 100 028-1 V1.4.1 (2001-12)
Measuring antenna Test antenna

Direct
Antenna to antenna coupling Path Loss Antenna to antenna coupling
Reflected
Path Loss

Imperfections
Mismatch + Mismatch
Attenuator 1 Attenuator 2
10 dB 10 dB
Mismatch + Mismatch
Cable Cable
Factor +
Factor
Measuring Test
antenna Cable loss Cable loss source
cable 1 cable 2
Mismatch + Mismatch
Measuring Test source

device Measuring device Signal generator
Figure 30: Schematic of the Transform Factor measurement
Measuring antenna: Input reflection coefficient: |ρRA| = 0,333
Measuring device: Input reflection coefficient: |ρRD| = 0,20
0,333 × 0,05 × 100

u j mismatch:antennaand attenuator2 = % = 1,177 %
2
ETSI
111 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,333 × 0,07 × 0,3162 × 100

u j cable1and antenna = % = 0,165 %
2
0,333 × 0,2 × 0,3162 × 0,8912 × 100

u j antennaand receiving device = % = 0,373 %
2
u j mismatch:measuring part = 1,1772 + 0,1652 + 0,3732 = 1,25%
Signal generator: absolute output level: There is not necessarily any similarity between the output levels from the
signal generator in the two stages of this test. As a result it contributes to both stages. The standard uncertainty of the
contribution, due to the signal generator absolute output level, is designated throughout all parts of TR 102 273 [3] as
uj38. Its value can be derived from manufacturer's data sheet.
uncertainty is taken from the manufacturers data sheet as ±1,0 dB. As nothing is said about the
ETSI
112 ETSI TR 100 028-1 V1.4.1 (2001-12)
cable is taken as 0,00 dB since in this measurement, the cable position, once set in this stage is not
subsequently changed during the test.
the two stages, a correction for measurement distance must be made to account for the different measurement distances.
In this test, once the position of the test antenna is set in stage one no further adjustment to its position is subsequently
made during the test. The standard uncertainty of the contribution due to the correction for measurement distance is
distance is taken as 0,00 dB.
EUT/substitution antenna. In this test, once the position of the test antenna is set in stage one no further adjustment to its
position is subsequently made during the test. The standard uncertainty of the contribution due to the correction for off
boresight angle in elevation plane is designated throughout all parts of TR 102 273 [3] as uj17.
angle in elevation plane is taken as 0,00 dB.
stages.
TR 102 273 [3] as uj46.
ETSI
113 ETSI TR 100 028-1 V1.4.1 (2001-12)
Ambient effect: Ambient effect: Ambient effect is the uncertainty caused by local ambient signals raising the noise
floor at the measurement frequency. The standard uncertainty of the contribution, due to the ambient effect, is
designated throughout all parts of TR 102 273 [3] as uj34. The values of the standard uncertainty should be taken from
table 27.
Receiving device noise floor (EUT OFF) is within: Standard uncertainty of

the contribution
coupling between the test antenna and its images in the absorbing material is designated throughout all parts of
TR 102 273 [3] as uj06.
test antenna and its images in the absorbing material is taken as 0,00 dB since this is the first stage of a
two stage measurement, where, in the second stage, the test antenna is located at the same height on the
mast as in this, the first stage.
input impedance and/or gain when placed close to a ground plane. The standard uncertainty of the contribution due to
the mutual coupling between the test antenna and its image in the ground plane is designated throughout all parts of
TR 102 273 [3] as uj14.
test antenna and its image in the ground plane is taken as 0,00 dB since this is the first stage of a two
stage measurement, where, in the second stage, the test antenna is located at the same height on the mast
as in this, the first stage.
Mutual coupling: measuring antenna to the test antenna: This is the effect produced by any change in gain of the
antennas which results from their close spacing. The standard uncertainty of the contribution due to the mutual coupling
of the measuring antenna to the test antenna, designated throughout all parts of TR 102 273 [3] as uj11, It has a standard
uncertainty of 0,00 dB for ANSI dipoles since it is included, where significant, in the mutual coupling and mismatch
loss correction factors. For non-ANSI dipoles the standard uncertainty can be taken from table 28.
ETSI
114 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 28: Uncertainty contribution of the mutual coupling between the measuring and test antenna

(3 m range) range)
measuring and test antennas is taken as 0,00 dB as we are using ANSI dipoles.
The standard uncertainty of the contribution, due to the interpolation of mutual coupling and mismatch loss correction
factors, is designated throughout all parts of TR 102 273 [3] as uj12. It has, for spot frequencies given in table A.20, a
value of 0,00 dB. However, for all other frequencies, the standard uncertainty should be obtained from table 29.
Table 29: Uncertainty contribution of the interpolation of mutual coupling

and mismatch loss correction factors
Frequency (MHz) Standard uncertainty of

the contribution
NOTE 5: In this example case the standard uncertainty of the contribution, due to the interpolation of mutual
coupling and mismatch loss correction factors, is taken as 0,00 dB as the frequency is above 180 MHz.
Range length: This contribution is associated with the curvature of the phase front from the measuring antenna to the
test antenna. The standard uncertainty of the contribution, due to range length, is designated throughout all parts of
TR 102 273 [3] as uj16. The standard uncertainty is 0,00 dB if ANSI dipoles are used. For other types of antenna the

2 2
1,26 dB
2 2
0,30 dB
2 2
0,10 dB
2
0,00 dB
NOTE 7: In this example case the standard uncertainty of the contribution, due to the range length, is taken as 0,00.
Reflectivity of absorbing material: measuring antenna to the test antenna: This uncertainty is the residual effect of
the reflections in the absorbing materials which result from the measuring antenna and EUT having different elevation
patterns. The standard uncertainty of the contribution due to the reflectivity of absorbing material between the
measuring and test antenna is designated throughout all parts of TR 102 273 [3] as uj02.
NOTE 8: In this example case the standard uncertainty of the contribution due to the reflectivity of absorbing
material between the measuring and test antenna is taken as 0,00 dB since a reference is set.
ETSI
115 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mutual coupling: measuring antenna to its images in the absorbing material: This uncertainty is the effect of the
change produced in the antenna's input impedance and/or gain. The standard uncertainty of the contribution, due to the
mutual coupling between the measuring antenna and its images in the absorbing material, is designated throughout all
measuring antenna and its images in the absorbing material is taken as 0,5 dB.
Mutual coupling: measuring antenna to its image in the ground plane: This uncertainty is the effect of the change
coupling of the measuring antenna to its image in the ground plane, is designated throughout all parts of
TR 102 273 [3] as uj14. Its value can be obtained from table 31.
Table 31: Uncertainty contribution of the mutual coupling between the measuring antenna
and its image in the ground plane

λ/2 ≤ spacing < 3λ/2 0,58 dB
measuring antenna to its image in the ground plane is taken as 0,15 dB since we are assuming vertical
polarization and a spacing of < 1,25 λ.
6.8.4.1.2.6 Measuring antenna
Antenna: antenna factor of the measuring antenna: Uncertainty is introduced as a result of the inaccurate knowledge
of the antenna factor of the measuring antenna. The antenna factor of the measuring antenna is only involved in the
second stage of the test and therefore does not act as a systematic offset. The standard uncertainty of the contribution,
due to the antenna factor of the measuring antenna, is designated throughout all parts of TR 102 273 [3] as uj44. For
ANSI dipoles the value should be obtained from table 32.
Table 32: Uncertainty contribution of the antenna factor of the measuring antenna.

contribution
the receiving antenna is 0,30 dB since ANSI dipoles have been used and the frequency is above
180 MHz.
Antenna: tuning of the measuring antenna Uncertainty is introduced as a result of the inaccurate tuning of the
measuring antenna. The standard uncertainty of the contribution due to the tuning of the measuring antenna is
designated in the present document by uj46.
NOTE 3: In this example case the standard uncertainty of the contribution due to the tuning of the measuring
antenna is taken to be 0,06 dB (see annex A) since the tuning of the measuring antenna is only involved in
this stage of the test and therefore does not act as a systematic offset.
ETSI
116 ETSI TR 100 028-1 V1.4.1 (2001-12)
Position of the phase centre: measuring antenna: Uncertainty is introduced as a result of the inaccurate positioning
of the phase centre of the measuring antenna, since it affects the range length i.e. the horizontal distance between itself
and the test antenna. The standard uncertainty of the contribution, due to the position of the phase centre of the
measuring antenna, is designated throughout all parts of TR 102 273 [3] as uj22.
the receiving antenna has been calculated from ± (the offset)/(range length) x 100 %). The positioning
uncertainty is ±0,01 m and therefore the worst case uncertainty = 0,01/3,0 = 0,333 %. As the offset can be
anywhere between these limits, the uncertainty is taken to be rectangularly distributed (see
TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) and the standard uncertainty is calculated as 0,192 %.
This is transformed to the logarithmic form (TR 100 028-2 [8], annex E), to be 0,02 dB.
6.8.4.1.2.7 Measuring antenna attenuator
Insertion loss: measuring antenna attenuator: The attenuator has an insertion loss as well as an uncertainty
the attenuator is not used outside the manufacturer's specification. At any given frequency the insertion loss is taken
from the manufacturer's data sheet since it does not cancel as a systematic offset (it only appears in one stage of the
test). The standard uncertainty of the contribution, due to the insertion loss uncertainty of the measuring antenna
attenuator, is designated throughout all parts of TR 102 273 [3] as uj40.
NOTE: In this example case the uncertainty of the contribution, due to the insertion loss uncertainty of the
measuring antenna attenuator, is taken from the manufacturer's data sheet as ±0,3 dB as nothing is said
about the distribution, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in
logs is assumed and the standard uncertainty is calculated as 0,17 dB.
6.8.4.1.2.8 Measuring antenna cable
Insertion loss: measuring antenna cable: The measuring antenna cable has an insertion loss as well as an uncertainty
contribution due to the insertion loss uncertainty of the measuring antenna cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
NOTE 1: In this example case the uncertainty of the contribution due to the insertion loss uncertainty of the
measuring antenna cable is taken from the manufacturer's data sheet as ±0,5 dB. As nothing is said about
the distribution, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is
assumed and the standard uncertainty is calculated as 0,29 dB.
Cable factor: measuring antenna cable: Cable factor is defined as the total effect of the antenna cable's influence on
inefficiency, parasitic effects (acting as a director or reflector) and introducing an unbalanced, common mode current
into the dipole balun. In a radiated measurement the standard uncertainty of the cable factor associated with the
measuring antenna cable is 0,5 dB provided the precautions detailed in the method have been observed i.e. routing and
dressing the cables with ferrites. If no prevention has been attempted the standard uncertainty is 4,0 dB (justification for
these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The standard uncertainty of the contribution, due
to the cable factor of the measuring antenna cable, is designated throughout all parts of TR 102 273 [3] as uj19.
NOTE 2: In the example case the standard uncertainty of the contribution due to the cable factor of the measuring
observed.
ETSI
117 ETSI TR 100 028-1 V1.4.1 (2001-12)
The receiving device is only used in the first stage of this test. Therefore, the absolute level uncertainty contributes fully
to this stage although the linearity does not.
Receiving device: absolute level: This uncertainty only contributes during the first stage of the measurement. The
standard uncertainty of the contribution due to the receiving device absolute level uncertainty is designated throughout
all parts of TR 102 273 [3] as uj47. Its value can be derived from manufacturers data.
NOTE 1: In this example case the uncertainty of the contribution due to the receiving device absolute level
uncertainty) is obtained from the manufacturers data as ±1 dB with a rectangular distribution (see
TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs. The standard uncertainty of the contribution due
to the receiving device absolute level uncertainty is calculated as 0,58 dB.
Receiving device: linearity: In any test in which the contribution of the absolute level uncertainty of the receiving
device contributes to the combined standard uncertainty of the test i.e. it does not cancel due to the methodology, the
contribution from the receiving device linearity is considered to have been included in uj47. Conversely, for any test in
which the absolute level uncertainty of the receiving device does not contribute to the combined standard uncertainty
the linearity of the receiving device should be included. The standard uncertainty of the contribution due to the
receiving device linearity is designated throughout all parts of TR 102 273 [3] as uj48.
taken as 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the Transform Factor. The
TR 102 273 [3] as ui01.
The Transform Factor measurement was repeated 10 times. The following results were obtained in dBµV (before
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3;
Y−
X2
32,10 × 10−6 −
(17,77 ×10 ) −3 2

n −1 10 − 1
is:
238,3 × 10 −6 100
−3
1,777 × 10 11,5
ETSI
118 ETSI TR 100 028-1 V1.4.1 (2001-12)

All the uncertainties for this part of the procedure are listed in table 33.
Table 33: Contributions from the transfer factor measurement

uj17 correction: off boresight angle in elevation plane , 0,00
uj11 mutual coupling: measuring antenna to the test antenna 0,00
uj02 reflectivity of absorber material: measuring antenna to the test antenna 0,00
uj06 mutual coupling: measuring antenna to its images in the absorbing material 0,50
uj14 mutual coupling: measuring antenna to its image in the ground plane 0,15
uj44 antenna: antenna factor of the measuring antenna 0,30
uj46 antenna: tuning of the measuring antenna 0,06
uj22 position of the phase centre: measuring antenna 0,02
uj40 insertion loss: measuring antenna attenuator 0,17
uj41 insertion loss: measuring antenna cable 0,29
uj19 cable factor: measuring antenna cable 0,50
sub-part 1, clause 5. This gives the combined standard uncertainty (uc Transform Factor) for the Transform Factor
measurement in dB.
The value of uc Transform Factor is calculated as 1,67 dB.

The second stage of the measurement (EUT sensitivity measurement) is to determine the minimum signal generator
output level which produces the required response from the EUT and converting the output level of the signal generator,
to a field strength using the Transform Factor derived in stage one. The test equipment set-up is shown in figure 31. The
components shown shaded are common to both stages of the test.
ETSI
119 ETSI TR 100 028-1 V1.4.1 (2001-12)
Test antenna
Test
antenna
Attenuator 2 cable 2 Signal
EUT generator
10 dB
ferrite beads
Ground plane
Figure 31: EUT measurement
Whereas figure 31 shows, schematically, the test equipment for the EUT sensitivity measurement, figure 32 an analysis
diagram, provides a detailed picture of the individual uncertainty components (each of which contributes its own
uncertainty) for this stage in the measurement. As stated above, the shaded areas represent components common to both
stages of the test method.

Mismatch: transmitting part: Only the transmitting part of the test equipment set-up is involved in this stage of the
test. The standard uncertainty of the contribution due to mismatch is 0,00 dB since, as stated in clause 6.8.4.1.1 the
transmitting part is common to both stages of the test.
Site factors Test antenna Mismatch
Antenna to Mutual
Antenna
EUT image +
gain
coupling coupling
Attenuator 2
10 dB
Direct Reflected
path loss path loss
Mismatch +
Site Cable
effects factor +
Test
EUT SINAD antenna Cable
meter loss
cable 2
Equipment
under test Mismatch +
Signal Signal
generator
generator
Figure 32: Reference for the measurement on the equipment (sensitivity)
ETSI
120 ETSI TR 100 028-1 V1.4.1 (2001-12)
The mismatch uncertainty between the signal generator and the test antenna contributes equally to both stages of the test
and therefore it does not contribute to the combined standard uncertainty and is not calculated. The standard uncertainty
of the contribution due to mismatch in the transmitting part is designated throughout all parts of TR 102 273 [3] as uj35.
NOTE: In this example case the standard uncertainty of the contribution due to mismatch in the transmitting part
is taken as 0,00 dB, since the uncertainty is systematic i.e. it produces the same offset in both stages.
Signal generator: absolute output level: There is not necessarily any similarity between the output levels from the
signal generator in the two stages of this test. As a result it contributes to both stages. The standard uncertainty of the
contribution, due to the signal generator absolute output level, is designated throughout all parts of TR 102 273 [3] as
uj38. Its value can be derived from manufacturer's data sheet.
uncertainty is taken from the manufacturers data sheet as ±1,0 dB. As nothing is said about the
insertion loss uncertainty of the test antenna cable is designated throughout all parts of TR 102 273 [3] as uj41.
cable is taken as 0,00 dB since in this measurement, the cable position, once set in stage 1, is not
subsequently changed during the test.
ETSI
121 ETSI TR 100 028-1 V1.4.1 (2001-12)
Insertion loss: test antenna attenuator: The test antenna attenuator has an insertion loss as well as an uncertainty
the attenuator is not used outside the manufacturer's specification. At any given frequency the insertion loss acts as a
contribution due to the insertion loss of the test antenna attenuator is designated throughout all parts of TR 102 273 [3]
as uj40.
the two stages, a correction for measurement distance must be made to account for the different measurement distances.
In this test, once the position of the test antenna is set in stage one no further adjustment to its position is subsequently
made during the test. The standard uncertainty of the contribution due to the correction for measurement distance is
distance is taken as 0,00 dB.
EUT/substitution antenna. In this test, once the position of the test antenna is set in stage one no further adjustment to its
position is subsequently made during the test. The standard uncertainty of the contribution due to the correction for off
boresight angle in elevation plane is designated throughout all parts of TR 102 273 [3] as uj17.
angle in elevation plane is taken as 0,00 dB since the uncertainty is assumed to be systematic i.e. it is
assumed constant and common to both stages of the measurement.
Antenna: gain of the test antenna: The gain of the test antenna acts as a systematic offset since it is present in both
stages of the test. The standard uncertainty of the contribution due to the gain of the test antenna is designated
NOTE 3: In this example case the standard uncertainty of the contribution due to the gain of the test antenna is
taken as 0,00 dB since the uncertainty is assumed to be systematic i.e. it is assumed constant and common
to both stages of the measurement.
The standard uncertainty of the contribution due to the tuning of the test antenna is designated throughout all parts of
TR 102 273 [3] as uj46.
taken as 0,00 dB since the uncertainty is assumed to be systematic i.e. it is assumed constant and common
to both stages of the measurement (provided that once set in stage one of the test it is not subsequently
re-tuned).
length. Since the test antenna position on the mast is not changed after being set in this stage the standard uncertainty of
the contribution due to the position of the phase centre of the test antenna is designated throughout all parts of
TR 102 273 [3] as uj22.
the test antenna is taken as 0,00 dB since the uncertainty is assumed to be systematic i.e. it is assumed
constant and common to both stages of the measurement.
ETSI
122 ETSI TR 100 028-1 V1.4.1 (2001-12)
measurement frequency. The standard uncertainty of the contribution due to the ambient effect is designated throughout
all parts of TR 102 273 [3] as uj34. The values of the standard uncertainties for this part of the test should be the same as
for stage 1.
0,00 dB as this was the value in stage 1.
Range length: This contribution is associated with the curvature of the phase front from the EUT to the test antenna.
The standard uncertainty of the contribution, due to range length, is designated throughout all parts of TR 102 273 [3]
as uj16. The standard uncertainty is 0,00 dB if ANSI dipoles are used. For other types of antenna the standard
uncertainty of the contribution should be obtained from table 34.

2 2
1,26 dB
2 2
0,30 dB
2 2
0,10 dB
2
0,00 dB
NOTE 3: In this example case the standard uncertainty of the contribution due to the range length is taken as
0,00 dB since the range length is ≥ 2 (d1 + d2)2/λ.
Mutual coupling: amplitude effect of the test antenna on the EUT: This uncertainty results from the interaction
between the EUT and the test antenna when placed close together. The standard uncertainty of the contribution due to
the amplitude effect of the mutual coupling between the test antenna and the EUT, is designated throughout all parts of
TR 102 273 [3] as uj08. The standard uncertainty should be taken from table 35.
Table 35: Uncertainty contribution: Mutual coupling: amplitude effect of the test antenna on the EUT
Range length Standard uncertainty of the

contribution
0,62√((d1 + d2) /λ)≤ range length < 2(d1 + d2) /λ
3 2
0,50 dB
2
0,00 dB
NOTE 4: In this example case the standard uncertainty of the contribution due to the amplitude effect of the mutual
coupling between the test antenna and the EUT is 0,00 dB since the distance is ≥ 2 (d1 + d2)2/λ.
Mutual coupling: EUT to its images in the absorbing material: This uncertainty is dependant on the quality of the
absorbing material and the effect imaging of the EUT in the ceiling, side and end walls has on the input impedance
and/or gain of the integral antenna. The standard uncertainty of the contribution, due to the mutual coupling of the EUT
to its images in the absorbing material, is designated throughout all parts of TR 102 273 [3] as uj04.
NOTE 5: In this example case the standard uncertainty of the contribution due to the mutual coupling amplitude
effect of the absorbing material on the EUT is assumed to be 0,5 dB (see annex A).
Mutual coupling: EUT to its image in the ground plane: This uncertainty results from the change in the EUT
spurious emission level as a result of being placed close to the ground plane. The standard uncertainty of the
contribution, due to the mutual coupling of the EUT to its image in the ground plane, is designated throughout all parts
of TR 102 273 [3] as uj13. Its value can be obtained from table 36.
ETSI
123 ETSI TR 100 028-1 V1.4.1 (2001-12)
Table 36: Uncertainty contribution of the mutual coupling

between the EUT to its image in the ground plane
Spacing between the EUT Standard uncertainty

For a vertically polarized EUT
For a horizontally polarized EUT
λ/2 ≤ spacing < 3λ/2 0,58 dB
NOTE 6: In this example case the standard uncertainty of the contribution, due to the mutual coupling of the EUT
to its image in the ground plane, (assuming the polarization is vertical and the spacing above the ground
plane is < 1,25 λ at the test frequency) is taken as 0,15 dB.
Reflectivity of absorbing material: EUT to the test antenna. This uncertainty is associated with the magnitudes of
the reflections occurring from the side walls, end walls and ceiling. These magnitudes are a function of the quality of
the absorber at the frequency of test. The standard uncertainty of the contribution, due to the reflectivity of the
absorbing material between the EUT and the test antenna, is designated throughout all parts of TR 102 273 [3] as uj01.
The relevant value for this contribution should be taken from table 37.
Table 37: Uncertainty contribution of the reflectivity

of absorbing material between the EUT and test antenna
Reflectivity of the Standard uncertainty of the

absorbing material contribution
reflectivity < 10 dB 4,76 dB
material between the EUT and the test antenna is taken as 1,24 dB.
coupling between the test antenna and its images in the absorbing material, is designated throughout all parts of
TR 102 273 [3] as uj06.
test antenna and its images in the absorbing material is taken as 0,00 dB since in this, the second stage of
the measurement, the uncertainty will only contribute if the test antenna is located at a different height to
the first stage which is not allowed by the methodology hence the uncertainty is assumed to be systematic
i.e. it is assumed constant and common to both stages of the measurement.
input impedance and/or gain when placed close to a ground plane. The standard uncertainty of the contribution, due to
the mutual coupling of the test antenna to its image in the ground plane, is designated throughout all parts of
TR 102 273 [3] as uj14.
antenna to its image in the ground plane is taken as 0,00 dB since in this, the second stage of the
measurement, the uncertainty will only contribute if the test antenna is located at a different height to the
first stage which is not allowed by the methodology hence the uncertainty is assumed to be systematic
i.e. it is assumed constant and common to both stages of the measurement.
ETSI
124 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.4.2.2.6 EUT
the precautions detailed in the methods have been observed. i.e. routing and dressing of cables with ferrites. If no
prevention has been attempted the standard uncertainty is 2,0 dB. The standard uncertainty of the contribution, due to
the mutual coupling of the EUT to the power leads, is designated throughout all parts of TR 102 273 [3] as uj54.
NOTE 1: In this example case the standard uncertainty of the contribution due to the mutual coupling of the EUT to
the power leads is taken as 0,5 dB since the precautions detailed in the methods have been observed.
Position of the phase centre: within the EUT volume: This contribution is associated with the uncertainty with which
the actual radiating point within the equipment volume is known. If this point is known exactly the contribution is
0,00 dB. The standard uncertainty of the contribution due to the position of the phase centre within the EUT volume is
NOTE 2: In this example case the standard uncertainty of the contribution due to the position of the phase centre
within the EUT volume has been calculated from (±(the maximum dimension of device)/(2 × range
length) × 100 %). In this example the position is not known. Hence, the uncertainty of the position of the
phase centre within the EUT of 0,15 m maximum dimension is 0,15/2 m = 0,075 m, and the worst case
uncertainty due to this offset is therefore (0,075/3,0) × 100 % = ±2,50 %. As the phase centre can be
anywhere inside the EUT, the uncertainty is taken as rectangularly distributed and the standard
uncertainty is calculated as 1,44 %. This is then transformed to the logarithmic form
(1,44/11,5) = 0,12 dB (TR 100 028-2 [8], annex E).
Positioning of the phase centre: within the EUT over the axis of rotation of the turntable: This contribution is
associated with the uncertainty with which the actual radiating point within the equipment is placed over the centre of
the turntable. If the point is placed exactly, the contribution is 0,00 dB. The standard uncertainty of the contribution, due
to the positioning of the phase centre within the EUT over the axis of rotation of the turntable, is designated throughout
all parts of TR 102 273 [3] as uj21.
NOTE 3: In this example case the standard uncertainty of the contribution due to the positioning of the EUT phase
centre over the axis of rotation of the turntable is calculated from (±(the estimated offset from the axis of
rotation)/(2 × range length) × 100 %). In this case, the uncertainty is taken as ±0,01 m, and the worst case
uncertainty is ±(0,01/3,0) × 100 = ±0,333 %. As the offset can be anywhere between the limits the
uncertainty is rectangularly distributed and the standard uncertainty is calculated as 0,192 %. This is then
transformed to the logarithmic form (0,192/11,5) = 0,02 dB (TR 100 028-2 [8], annex E).
EUT: degradation measurement: This contribution is a RF level uncertainty associated with the uncertainty of
measuring 20 dB SINAD, 10-2 bit stream or 80 % message acceptance ratio. The standard uncertainty of the
contribution, due to the EUT degradation measurement, is designated throughout all parts of TR 102 273 [3] as uj52. Its
value can be obtained from TR 100 028 (all parts).
NOTE 4: In this example case, the standard uncertainty of the contribution is obtained from TR 100 028 (all parts)
and its value is 0,68 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the receiver sensitivity. The
TR 102 273 [3] as ui01.
The receiver sensitivity measurement was repeated 10 times. The following results were obtained in dBµV (before
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3.
ETSI
125 ETSI TR 100 028-1 V1.4.1 (2001-12)
Y−
X2
32,10 × 10 −6
−
(17,77 ×10 )−3 2

n −1 10 − 1
is:
238,3 ×10 −6 100

−3
1,777 ×10 11,5

All the uncertainty contributions for this part of the procedure are listed in table 38.

uj04 Mutual coupling: EUT to its images in the absorbing material 0,50
uj13 mutual coupling: EUT to its image in the ground plane 0,15
uj01 reflectivity of absorber material: EUT to the test antenna 1,24
uj20 position of the phase centre: within the EUT volume 0,12
uj22 positioning of the phase centre: within the EUT over the axis of rotation of the turntable 0,02
uj52 EUT: degradation measurement 0,68
sub-part 1, clause 5. This gives the combined standard uncertainty (uc EUT measurement) for the NSA measurement in dB.
The value of uc EUT measurement is calculated as 2,06 dB.
ETSI
126 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.8.4.2.5 Expanded uncertainty for the receiver Sensitivity measurement

outlined in clauses 6.8.2.1.4 and 6.8.2.2.4. The components to be combined are uc Transform Factor and uc EUT measurement.
uc Sensitivity measurement = 1,67 2 + 2,062 = 2,65 dB
6.9 Examples of measurement uncertainty analysis (Stripline)

6.9.1 Introduction
This clause contains detailed worked examples of the calculation of expanded uncertainty of radiated tests in a Stripline
test facility.
The example tests given are limited to:
- a verification procedure;
- the measurement of a receiver parameter (Sensitivity).
Both of the example tests are assumed to have been carried out in the 2-plate open Stripline described in CENELEC
European Standard EN 55020 [7].
NOTE 1: The values given to all of the uncertainty components in these examples are only to illustrate the
uncertainty calculation. In practice, values should be derived by consulting annex A.
All radiated tests consist of two stages. For the verification procedure, the first stage is to set a reference level whilst the
second stage involves the measurement of path loss/attenuation through the Stripline. For the measurement of the
receive sensitivity of an EUT, the first stage is to find the minimum (or average) signal generator output that produces
the required receiver response, whilst the second stage determines the field strength in the Stripline corresponding to
that generator output level.
Within any radiated test there are uncertainty components that are common to both stages of the test. By their different
natures some of these common uncertainties will cancel (e.g. the uncertainty of the insertion loss of a cable common to
both parts), others will contribute once (e.g. the increase in the field strength caused by a large EUT) whilst others may
contribute twice (e.g. the level stability of a signal generator in receiver tests). In each of the following uncertainty
analyses, uncertainty components common to both stages are shown as shaded areas in the accompanying schematic
diagrams.
As shown in the examples, all the individual uncertainty components for any test are combined in the manner described
in TR 102 273 [3], part 1, sub-part 1, clauses 4 and 5 in order to derive an expanded uncertainty figure for the
measurement. The values of the individual components are either provided in annex A or should be taken from
manufacturers' data sheets. Whenever they are obtained from data sheets, worst case figures given over a frequency
band should be used. For example, in the case of a signal generator whose absolute level accuracy is quoted as ±1 dB
over 30 MHz to 300 MHz, ±2 dB over 300 MHz to 1 000 MHz the figure for the band containing the test frequency
should be used. This approach should be adopted for all uncertainty components, taking the uncertainty figures over as
broad a band a possible. This is normally satisfactory when the variation with frequency is not large and provides a
simple and flexible approach. The resulting expanded uncertainty figure is valid across a broad range of frequencies and
measurement conditions and avoids the necessity of repeated calculation for minor frequency changes.
NOTE 2: Taking specific frequency values may result in a lower expanded uncertainty value, but this lower value is
only valid when that specific set of circumstances apply for which the value was derived.
6.9.2 Example 1: Verification procedure

The Stripline verification procedure involves two different measurement stages and it results in values for both the
attenuation through the Stripline and its Transform Factor (i.e. the relationship between the input voltage to the Stripline
and the resulting field strength between the plates).
ETSI
127 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.2.1 Uncertainty contributions: Stage 1: Direct attenuation measurement

The first stage of the verification procedure is the direct attenuation measurement. This is carried out with all the items
of test equipment connected directly together via an "in line" adapter between the attenuators as shown in figure 33. The
components shown shaded are common to both stages of the procedure.
cable 1 cable 2
Signal Attenuator 1 "In line" Attenuator 2 Receiving
generator 10 dB adapter 10 dB device
Figure 33: Stage 1: Direct attenuation measurement
Despite the commonality of most of the components to both stages of this procedure, the mismatch uncertainty
contribution for both stages has to be calculated and included in the uncertainty calculations. This is the result of load
conditions varying (i.e. the Stripline and monopole replaces the adapter in the second stage). Conversely, as a result of
this commonality, the uncertainty contributions of some of the individual components will cancel.

Mismatch: direct attenuation measurement: The value of the combined standard uncertainty of the contribution due
to the mismatch from the source to the receptor, i.e. between the signal generator and the receiving device, is calculated
from the approach described in annex G. All the individual contributions are U-distributed.
NOTE 1: In this example the signal generator output reflection coefficient used is the worst case magnitude over
the frequency band of interest, as is the case with the cable, adapter and attenuator VSWRs.
NOTE 2: The attenuation values of the cables and attenuators should be obtained from the manufacturers data
sheet/calibration records at the specific frequency of the test, along with the associated uncertainties for
these values.
ETSI
128 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mismatch Mismatch
Attenuator 1 + Attenuator 2
+ Adapter
10 dB 10 dB
Cable Cable
+ +
factor factor
antenna Cable Cable antenna
cable 1 loss loss cable 2
source Receiving
device
Signal
generator device
Figure 34: Stage 1: Direct attenuation measurement individual uncertainty components
Adapter: Input and output reflection coefficients |S11| = |S22| = 0,02
Attenuation = 0,1 dB |S12| = |S21| = 0,9886
ETSI
129 ETSI TR 100 028-1 V1.4.1 (2001-12)
Mismatch uncertainty in the direct attenuation measurement: In the following the transmitting antenna cable is
named cable 1, the transmitting antenna attenuator is named attenuator 1, the receiving antenna cable is named cable 2,
the receiving attenuator is named attenuator 2. Those components that are constant for both stages 1 and 2 are not
calculated as they do not contribute to the overall uncertainty.
Mismatch between:
0,05 × 0,02 × 100

u j mismatch:attenuator1and adapter = % = 0,071 %
2
0,02 × 0,05 × 100

u j mismatch:adapter and attenuator1 = % = 0,071 %
2
0,07 × 0,07 × 0,3162 × 100

u j mismatch:cable1and adapter = % = 0,035 %
2
0,05 × 0,05 × 0,9882 × 100

u j mismatch:attenuator1and attenuator2 = % = 0,173 %
2
0,02 × 0,07 × 0,3162 × 100

u j mismatch:adapter and cable2 = % = 0,010 %
2
0,2 × 0,02 × 0,8912 × 0,3162 × 100

u j mismatch:generatorand adapter = % = 0,022 %
2
0,07 × 0,05 × 0,3162 × 0,9882 × 100

u j mismatch:cable1and attenuator2 = % = 0,024 %
2
0,05 × 0,07 × 0,9882 × 0,3162 × 100

u j mismatch:attenuator1and cable2 = % = 0,024 %
2
0,02 × 0,2 × 0,3162 × 0,8912 × 100

u j mismatch:adapter and receiving device = % = 0,022 %
2
0,2 × 0,05 × 0,8912 × 0,3162 × 0,9882 × 100

u j mismatch:generatorand attenuator2 = % = 0,055 %
2
- uj mismatch: cable 1 and cable 2: Less than 0,01 % due to the two attenuators, therefore neglected.
0,05 × 0,2 × 0,988 2 × 0,316 2 × 0,8912 × 100

u j mismatch:attenuator1and receiving device = % = 0,055 %
2
ETSI
130 ETSI TR 100 028-1 V1.4.1 (2001-12)
- uj mismatch: generator and cable 2: Less than 0,01 % due to the two attenuators, therefore neglected.
- uj mismatch: cable 1 and receiving device: Less than 0,01 % due to the two attenuators, therefore neglected.
- uj mismatch: generator and receiving device: Less than 0,01 % due to the two attenuators, therefore neglected.
uc mismatch:direct att. = 0,0712 + 0,0712 + ... + 0,0552 + 0,0552 = 0,306 %
Transforming to logarithmic form (TR 100 028-2 [8], annex E): 0,306 %/11,5 = 0,026 dB.
6.9.2.1.2.2 Signal generator cable
Insertion loss: signal generator cable: The signal generator cable has an insertion loss as well as an uncertainty
contribution due to the insertion loss uncertainty of the signal generator cable is designated throughout all parts of
TR 102 273 [3] as uj41.
ETSI
131 ETSI TR 100 028-1 V1.4.1 (2001-12)
Cable factor: signal generator cable: Cable factor is defined as the total effect of the signal generator cable's influence
on the measuring system including its interaction with the Stripline. It consists of the leakage caused by cable screening
inefficiency and introducing an unbalanced current into the Stripline. In a radiated measurement the standard
uncertainty of the cable factor, associated with each cable, is 0,5 dB provided that the precautions detailed in the
methods have been observed. i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the
standard uncertainty is 4,0 dB (justification for these values is given in TR 102 273 [3], part 1, sub-part 2, annex E). The
standard uncertainty of the contribution due to the cable factor of the signal generator cable is designated throughout all
NOTE 2: In this example case the standard uncertainty of the contribution due to the cable factor of the signal
generator is taken as 0,00 dB since there are no external fields involved other than leakage, which is
6.9.2.1.2.3 Signal generator attenuator
Insertion loss: signal generator attenuator: The signal generator attenuator has an insertion loss as well as an
the contribution, due to the insertion loss uncertainty of the signal generator attenuator, is designated throughout all
the signal generator attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
6.9.2.1.2.4 Adapter
Insertion loss: adapter: The adapter has an insertion loss as well as an uncertainty associated with the measurement of
its magnitude. The value of insertion loss and its uncertainty remain valid provided the adapter is not used outside the
manufacturer's specification. The standard uncertainty of the contribution due to the insertion loss uncertainty of the
adapter is designated throughout all parts of TR 102 273 [3] as uj42. Its value can be derived from the manufacturer's
data sheet.
NOTE: In this example case the uncertainty of the contribution due to the insertion loss uncertainty of the adapter
is obtained from the manufacturers data sheet as ±0,10 dB. As nothing is said about the distribution of
this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is
6.9.2.1.2.5 Receiving device attenuator
Insertion loss: receiving device attenuator: The attenuator has an insertion loss as well as an uncertainty associated
contribution due to the insertion loss uncertainty of the receiving device attenuator is designated throughout all parts of
TR 102 273 [3] as uj40.
the receiving device attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
ETSI
132 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.2.1.2.6 Receiving device cable
Insertion loss: receiving device cable: The receiving device cable has an insertion loss as well as an uncertainty
contribution, due to the insertion loss uncertainty of the receiving device cable, is designated throughout all parts of TR
102 273 [3] as uj41.
the receiving device cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same
Cable factor: receiving device cable: Cable factor is defined as the total effect of the receiving device cable's
influence on the measuring system including its interaction with the Stripline. It consists of the leakage caused by cable
screening inefficiency and introducing an unbalanced current into the Stripline. In a radiated measurement the standard
methods have been observed i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the
standard uncertainty of the contribution due to the cable factor of the receiving device cable is designated throughout all
device cable is taken as 0,00 dB since there are no external fields involved other than leakage, which is
In this, the first stage of the Stripline verification procedure, a reference level is set on the receiving device for a
particular output level from the signal generator. In the second stage (where the path loss through the Stripline is
measured), a second level is obtained on the receiving device. Only in the second stage do the linearity and absolute
level uncertainties of the receiver become involved in the calculation of the combined standard uncertainty for the
measurement.
Receiving device: absolute level: The standard uncertainty of the contribution due to the receiving device absolute
level uncertainty is designated throughout all parts of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution due to the absolute level uncertainty of
the receiving device is assumed to be 0,00 dB since, in this part of the measurement, a reference level is
recorded.
Receiving device: linearity: The standard uncertainty of the contribution due to the receiving device linearity is
assumed to be 0,00 dB since, in this part of the measurement, a reference level is recorded.

Random uncertainty: The magnitude can be assessed from multiple measurements of the direct attenuation
parts of TR 102 273 [3] as ui01. Its value can then be calculated.
The direct attenuation measurement was repeated 10 times. The following results (dBµV) in were obtained (before
- 106,8; 107,2; 106,7; 107,0; 107,2; 106,7; 107,1; 106,8; 107,1; 107,0.
- 0,2188; 0,2291; 0,2163; 0,2239; 0,2291; 0,2163; 0,2265; 0,2188; 0,2265; 0,2239.
ETSI
133 ETSI TR 100 028-1 V1.4.1 (2001-12)
- X = the sum of the measured values = 2,2292 V;
- Y = the sum of the squares of the measured values = 0,4972 V2.
X2 2,2292 2
Y− 0,4972 −
n −1 10 − 1
is:
5,444 × 10−3 100

0,22292 11,5
0,212 dB.

A complete list of all the contributions to this part of the verification procedure is given in table 39.
Table 39: Contributions from the reference, direct measurement

uj35 mismatch: direct attenuation measurement 0,03
uj41 insertion loss: signal generator cable 0,00
uj19 cable factor: signal generator cable 0,00
uj40 insertion loss: signal generator attenuator 0,00
uj42 insertion loss: adapter 0,06
uj40 insertion loss: receiving device attenuator 0,00
uj41 insertion loss: receiving device cable 0,00
uj19 cable factor: receiving device cable 0,00
sub-part 1, clause 5. This gives the combined standard uncertainty (uc direct attenuation measurement) for the direct
The value of uc direct attenuation measurement is calculated as 0,223 dB.
6.9.2.2 Uncertainty contributions: Stage 2: Radiated attenuation measurement

The second stage of the verification procedure is the Stripline radiated attenuation measurement. This involves
mounting a monopole antenna through a hole in the lower plate of the Stripline, so that the feed point to the monopole is
flush with the surface of the lower plate. The radiated attenuation measurement is carried out by removing the adapter
and connecting the signal generator attenuator to the Stripline and connecting the receiving device attenuator to the
monopole output (figure 35). The difference in received levels (after allowance for any correction factors which may be
appropriate), for the same signal generator output level, is the Stripline radiated attenuation.
ETSI
134 ETSI TR 100 028-1 V1.4.1 (2001-12)
Central axis
of stripline
Signal Load
generator
Ferrite beads 150 Ω

10dB termination
attenuator
Receiving
device
ETSI
135 ETSI TR 100 028-1 V1.4.1 (2001-12)

Mismatch in the transmitting and receiving parts: The value of the combined standard uncertainty of the
contributions due to the mismatch are calculated from the approach described in annex G. It is calculated in two parts.
Firstly the standard uncertainty of the contribution due to the mismatch in the transmitting part, i.e. between the signal
generator, cable, attenuator and the input to the Stripline and secondly, that for the receiving part, i.e. between the
receiving monopole antenna, attenuator, cable and the receiving device.
Stripline
Stripline
Radiated
to monopole Monopole
Loss
coupling
Attenuator 1
Site Attenuator 2
10 dB effects 10 dB
Room
resonances, etc.
Cable Cable
+ +
factor factor
Signal Receiving
generator Cable Cable device
loss loss
cable 1 cable 2
Signal Signal Receiving Receiving

generator device
generator device
All the individual contributions are U-distributed.
NOTE 1: In this example value taken for the signal generator output reflection coefficient is the worst case over the
frequency band of interest, similarly, for the cable, adapter and attenuator VSWRs.
ETSI
136 ETSI TR 100 028-1 V1.4.1 (2001-12)
Stripline: Input reflection coefficient: |ρSL| = 0,333
Those components that cancel are not calculated. Other contributions are (see annex G):
0,05 × 0,333 × 100

u j mismatch:attenuator1and Stripline = % = 1,177 %
2
0,07 × 0,333 × 0,3162 × 100

u j mismatch:cable1and Stripline = % = 0,165 %
2
0,2 × 0,333 × 0,8912 × 0,3162 × 100

u j mismatch:generatorand Stripline = % = 0,373 %
2
uc mismatch:transmitting part = 1,1772 + 0,1652 + 0,3732 = 1,25%
transforming to the logarithmic form (TR 100 028-2 [8], annex E): 1,25 %/11,5 = 0,11 dB.
Monopole: Input reflection coefficient: |ρM| = 0,333
Receiving device: Output reflection coefficient: |ρRD| = 0,20
0,333 × 0,05 × 100

u j mismatch:monopoleand attenuator = % = 1,177 %
2
ETSI
137 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,333 × 0,07 × 0,3162 × 100

u j mismatch:monopoleand cable2 = % = 0,165 %
2
0,333 × 0,2 × 0,3162 × 0,8912 × 100

u j mismatch:monopoleand receiving device = % = 0,373%
2
uc mismatch:receiving part = 1,1772 + 0,1652 + 0,3732 = 1,25%
The standard uncertainty of the contribution due to the mismatch in the receiving part is designated throughout all parts
of TR 102 273 [3] as uj37. Its value in this example is 0,11 dB.
contribution, due to the insertion loss uncertainty of the signal generator cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
the signal generator cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same
ETSI
138 ETSI TR 100 028-1 V1.4.1 (2001-12)
on the measuring system including its interaction with the Stripline. It consists of the leakage caused by cable screening
observed.
the contribution, due to the insertion loss uncertainty of the signal generator attenuator, is designated throughout all
the signal generator attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the

0,00 dB since the Stripline is assumed to be placed in a shielded room.
Stripline: influence of site effects: The influence of site effects comprise those effects resulting from not observing the
recommendations given in EN 55020 [7] regarding positioning of the Stripline and layout of the absorber. These can
lead to incorrect received levels i.e. values which differ from theoretical calculations. The standard uncertainty of the
contribution due to the influence of site effects is designated throughout all parts of TR 102 273 [3] as uj33.
NOTE 2: In this example case the standard uncertainty of the contribution due to the influence of site effects is
taken to have a standard uncertainty of 3,0 dB (see annex A). In the verification procedure, site effects
only contribute to the Stripline radiated attenuation part.
ETSI
139 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.2.2.2.5 Antenna factor of the monopole
Stripline: antenna factor of the monopole: This has been derived from measurements taken within the Stripline.
Therefore, the given values incorporate several of the field disturbance factors which the Stripline possesses and which
therefore do not have to be allowed for as individual contributions. Amongst these included effects are imaging,
characteristic impedance of the line, non-planar nature of the field etc. The standard uncertainty of the contribution due
to the antenna factor of the monopole is designated throughout all parts of TR 102 273 [3] as uj30.
NOTE: In this example case the standard uncertainty of the contribution due to the antenna factor of the
monopole is taken as 1,15 dB. This combined uncertainty source is only present in the Stripline radiated
attenuation measurement.
6.9.2.2.2.6 Receiving device attenuator
Insertion loss: receiving device attenuator: The receiving device attenuator has an insertion loss as well as an
the contribution due to the insertion loss uncertainty of the receiving device attenuator is designated throughout all parts
of TR 102 273 [3] as uj40.
the receiving device attenuator is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the
contribution, due to the insertion loss uncertainty of the receiving device cable, is designated throughout all parts of
TR 102 273 [3] as uj41.
the receiving device cable is taken as 0,00 dB since the uncertainty is systematic i.e. it produces the same
influence on the measuring system including its interaction with the site. It consists of the leakage caused by cable
screening inefficiency and introducing an unbalanced current into the monopole. In a radiated measurement the
standard uncertainty of the cable factor, associated with each cable, is 0,5 dB provided that the precautions detailed in
the methods have been observed i.e. routing and dressing of cables with ferrites. If no prevention has been attempted the
standard uncertainty of the contribution due to the cable factor of the receiving device cable is designated throughout all
antenna cable is taken as 0,50 dB since the precautions detailed in the methods have been observed.
The first stage of the verification procedure involved setting a reference level on the receiving device for a particular
output level from the signal generator. In this the second stage (where the radiated attenuation through the Stripline is
measured), a second level is obtained which results in linearity and absolute level uncertainties becoming involved in
the calculation of the combined standard uncertainty for the measurement.
of TR 102 273 [3] as uj47.
ETSI
140 ETSI TR 100 028-1 V1.4.1 (2001-12)
NOTE 1: In this example case the uncertainty of the contribution due to the receiving device absolute level
uncertainty (a range change is assumed) is obtained from the manufacturers data as ±0,5 dB. As nothing
is said about the distribution of this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1,
sub-part 1, clause 5.1.2) in logs is assumed. The standard uncertainty of the contribution due to the
receiving device absolute level uncertainty is calculated as 0,29 dB.
taken as 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the radiated attenuation
The radiated attenuation measurement was repeated 10 times. The following results were obtained in dBµV (before
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3;
Y−
X2
32,10 × 10 −6
−
(17,77 ×10 )−3 2

n −1 10 − 1
is:
238,3 ×10 −6 100

−3
1,777 ×10 11,5
evaluated as 1,17 dB. See also the note in clause 6.4.7.
ETSI
141 ETSI TR 100 028-1 V1.4.1 (2001-12)

A complete list of all the contributions to this part of the verification procedure is given in table 41.
Table 41: Contributions from the radiated attenuation measurement

uj33 Stripline: influence of site effects 3,00
uj31 Stripline: antenna factor of the monopole 1,15
uj40 insertion loss: receiving device attenuator 0,00
sub-part 1, clause 5. This gives the combined standard uncertainty (uc Stripline attenuation measurement) for the Stripline
The value of uc Stripline attenuation measurement is calculated as 3,51 dB.
6.9.2.2.5 Expanded uncertainty for the verification procedure

outlined in clauses 6.9.2.1.4 and 6.9.2.2.4. The components to be combined are uc direct attenuation measurement and
uc Stripline attenuation measurement.
uc Striplineverification procedure = 0,2232 + 3,512 = 3,51 dB
The expanded uncertainty is ±1,96 x 3,51 = ±6,89 dB at a 95 % confidence level.
6.9.3 Example 2: The measurement of a receiver parameter (Sensitivity)

For the measurement of receiver sensitivity two stages of test are involved. The first stage (determining the Transform
Factor of the Stripline) involves measuring the field strength at the point where the receiver will be placed and
determining the relationship between the signal generator output power level and the resulting field strength. The
second stage (the EUT measurement) involves placing the EUT within the Stripline and adjusting the output level of the
signal generator until the required response is obtained on the receiver. The signal generator output power level is then
converted to field strength using the Transform Factor.
ETSI
142 ETSI TR 100 028-1 V1.4.1 (2001-12)

The first stage of the measurement is to determine the minimum signal generator output level which produces the
required response from the EUT. The test equipment set-up is shown in see figure 37.
EUT with volume centre Central axis

of stripline 150 Ω
midway between plates
termination
Signal Load
generator
Ferrite beads Non-conducting,

10dB
Modulation low dielectric constant
attenuator detection support stand
Figure 37: Stage 1: EUT measurement
Whereas figure 37 shows, schematically, the test equipment set-up for the EUT sensitivity measurement, figure 38, an
analysis diagram, provides a detailed picture of the individual uncertainty components (each of which contributes its
own uncertainty) for this stage in the measurement. As stated above, the shaded areas represent components common to
both stages of the test method.

Mismatch: transmitting part: The uncertainty due to mismatch for the measurement on the EUT concerns only the
signal generator, the signal generator cable, the signal generator attenuator and the input to the Stripline. The mismatch
uncertainty through this network does, however, contribute equally to both stages of the test for cases in which a field
strength measurement is subsequently performed.
If, however, the results of the verification procedure are used to calculate the field strength, the contribution of the
mismatch uncertainty needs to be calculated, from the approach described in annex G. All the contributions are
U-distributed.
Stripline: Input reflection coefficient: |ρSL| = 0,333
ETSI
143 ETSI TR 100 028-1 V1.4.1 (2001-12)
Stripline
Radiated Stripline to
EUT EUT
loss
coupling
Mismatch +
Signal Site
Attenuator 1 Modulation
generator 10 dB effects detector
attenuator Room
resonances, etc.
Mismatch +
Cable
+
factor
Signal
Cable
generator loss
cable 1
Mismatch +
Signal
generator Signal
generator
Modulation
source
Figure 38: Schematic of the sensitivity measurement on the EUT
Mismatch: transmitting part.

annex G):
0,05 × 0,333 × 100

u j mismatch:attenuator1and Stripline = % = 1,177 %
2
ETSI
144 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,07 × 0,333 × 0,316 2 × 100

u j mismatch:cable1and Stripline = % = 0,165 %
2
0,2 × 0,333 × 0,8912 × 0,316 2 × 100

u j mismatch:generator and Stripline = % = 0,373 %
2
u cmismatch:transmitting part = 1,177 2 + 0,165 2 + 0,3732 = 1,25 %
Signal generator: absolute output level: In this test method, the uncertainty due to the setting of the signal generator's
absolute output level contributes to both stages. In stage 1, the output level is individually adjusted at each of 8 different
positioning angles whilst in stage 2, after an inspection (or calculation) of the 8 different values, the signal generator is
set to a specific output level. The standard uncertainty of the contribution due to the signal generator absolute output
level uncertainty is designated throughout all parts of TR 102 273 [3] as uj38. Its value can be obtained from the
manufacturer's data sheet.
NOTE 1: In this example case the uncertainty of the contribution due to the signal generator absolute output level is
TR 102 273 [3] as uj41.
the signal generator cable is taken as 0,20 dB as the results of the verification procedure have been used
to determine the field strength and this value has been taken from the manufacturer's or calibration data.
NOTE 2: If a field measurement had been performed using either a monopole of 3-axis probe the value would have
been 0,00 dB as the uncertainty is systematic i.e. it produces the same offset in both stages.
ETSI
145 ETSI TR 100 028-1 V1.4.1 (2001-12)
generator cable is taken as 0,5 dB since the precautions detailed in the test method have been observed.
the contribution due to the insertion loss of the signal generator attenuator is designated throughout all parts of
TR 102 273 [3] as uj40.
the signal generator cable is taken as 0,20 dB as the results of the verification procedure have been used
to determine the field strength and this value has been taken from the manufacturer's or calibration data.
TR 102 273 [3] as uj34. The value of the standard uncertainty is the same as for the second stage.
0,00 dB since the Stripline is assumed to have been placed in a shielded room.
Stripline: influence of site effects: The influence of site effects comprise those effects, resulting from not observing
the recommendations given in EN 55020 [7] regarding positioning of the Stripline and layout of absorber. These can
lead to incorrect received levels i.e. values which differ from theoretical calculations. In this example, the
recommendations are assumed not to have been fully observed. The standard uncertainty of the contribution due to the
influence of site effects is designated throughout all parts of TR 102 273 [3] as uj33.
taken as 3,0 dB since in the verification procedure, site effects only contribute to the Stripline attenuation
part and therefore do not cancel.
ETSI
146 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.3.1.2.5 EUT
Stripline: mutual coupling of the EUT to its images in the plates: The magnitude is dependent on the EUT's size.
The EUT is assumed to be positioned midway between the plates. The standard uncertainty of the contribution due to
the mutual coupling of the EUT to its images in the plates is designated throughout all parts of TR 102 273 [3] as uj24.
Its value can be obtained from table 42.
Table 42: Uncertainty contribution of the mutual coupling of the EUT to its images in the plates
Size of the EUT relative to the plate Standard uncertainty of the

separation contribution
size/separation < 33 % 1,15 dB
33 % ≤ size/separation < 50 % 1,73 dB
50 % ≤ size/separation < 70 % 2,89 dB
70 % ≤ size/separation ≤ 87,5 % (max.) 5,77 dB
its images in the plates is 1,15 dB since it is assumed the equipment size is < 33 % of the plate separation.
Stripline: characteristic impedance: This uncertainty contribution results from the difference between the free-space
wave impedance (377 Ω) for which the EUT had been developed and that for the Stripline (150 Ω). The standard
uncertainty of the contribution due to the characteristic impedance of the Stripline is designated throughout all parts of
TR 102 273 [3] as uj26.
NOTE 2: In this example case the standard uncertainty of the contribution due to the characteristic impedance of
the Stripline is taken as having a standard uncertainty of 0,58 dB.
Stripline: correction factor for the size of the EUT: This uncertainty is the result of changes in the intensity of the
electric field between the plates resulting from the presence, and metal content of the EUT. The larger the size of the
EUT in the vertical plane of the Stripline, the greater the field intensification effect. Correction factors are supplied
within the test method, and the associated standard uncertainty of the contribution of the uncertainty due to the
correction factor for the size of the EUT is designated throughout all parts of TR 102 273 [3] as uj32. For EUT mounted
centrally in the Stripline, values can be obtained from table 43.
Table 43: Uncertainty contribution: Stripline: correction factor for the size of the EUT
Height of the EUT Standard uncertainty of the

(in the E-plane) is: contribution
height < 0,2 m 0,30 dB
0,2 m ≤ height < 0,4 m 0,60 dB
0,4 m ≤ height ≤ 0,7 m 1,20 dB
NOTE 3: In this example case the standard uncertainty of the contribution of the uncertainty due to the correction
factor for the size of the EUT is taken as 0,60 dB as the EUT is 0,22 m high.
the precautions detailed in the methods have been observed. i.e. routing and dressing of cables with ferrites. If no
prevention has been attempted the standard uncertainty is 2,0 dB. The standard uncertainty of the contribution due to
the mutual coupling of the EUT to the power leads is designated throughout all parts of TR 102 273 [3] as uj54.
the power leads is taken as 0,5 dB since the precautions detailed in the methods have been observed.
Stripline: non-planar nature of the field distribution: This uncertainty results from the non-uniform amplitude and
phase distribution of the electric field across the EUT. The non-uniformity results from room resonances, constructional
problems, moding, reflections, etc. The standard uncertainty of the contribution due to the non-planar nature of the field
distribution is designated throughout all parts of TR 102 273 [3] as uj27.
NOTE 5: In this example case the standard uncertainty of the contribution due to the non-planar nature of the field
distribution is taken as 0,29 dB.
ETSI
147 ETSI TR 100 028-1 V1.4.1 (2001-12)
EUT: degradation measurement: This contribution is a RF level uncertainty associated with the uncertainty of
measuring 20 dB SINAD, 10-2 bit stream or 80 % message acceptance ratio. The standard uncertainty of the
contribution due to the EUT degradation measurement, designated throughout all parts of TR 102 273 [3] as uj52, can be
obtained from TR 100 028 (all parts).
NOTE 6: In this example case, the standard uncertainty of the contribution is obtained from TR 100 028 (all parts)
and its value is 0,68 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the receiver sensitivity
The receiver sensitivity measurement was repeated 10 times. The following results were obtained in dBµV (before
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3.
Y−
X2
32,10 × 10 −6
−
(17,77 ×10 )−3 2

n −1 10 − 1
is:
238,3 × 10 −6 100
−3
1,777 × 10 11,5
ETSI
148 ETSI TR 100 028-1 V1.4.1 (2001-12)

A complete list of all the contributions to this part of the test method is given in table 44.
Table 44: Contributions from the measurement on the EUT

uj36 mismatch: transmitting part:
a) Using results of the verification procedure; 0,11
b) Using a monopole for field measurement; 0,00
c) Using a 3-axis probe for field measurement. 0,00
uj41 insertion loss: signal generator cable:
c) Using 3-axis probe for field measurement. 0,00
uj40 insertion loss: signal generator attenuator:
uj33 Stripline: influence of site effects:
uj24 Stripline: mutual coupling of the EUT to its images in the plates 1,15
uj26 Stripline: characteristic impedance 0,58
uj32 Stripline: correction factor for the size of the EUT 0,60
uj27 Stripline: non-planar nature of the field distribution 0,29
uj52 EUT: degradation measurement 0,68
ui01 random uncertainty 1,17
sub-part 1, clause 5. This gives the combined standard uncertainty for the direct attenuation measurement in dB as
follows:
- using results of the verification procedure = uc measurement of the EUT = 3,72 dB;
- using a monopole for field measurement = uc measurement of the EUT = 2,18 dB;
- using a 3-axis probe for field measurement = uc measurement of the EUT = 2,18 dB.
6.9.3.2 Uncertainty contributions: Stage 2: Field measurement using the results of

the verification procedure
Stripline: interpolation of values for the Transform Factor: In this case, the frequency of test does not coincide with
a frequency at which the verification procedure was carried out. Therefore, a contribution is included to account for the
interpolation between Transform Factor values. The standard uncertainty of the contribution due to the interpolation of
values for the Transform Factor of the Stripline is designated throughout all parts of TR 102 273 [3] as uj30.
NOTE: In this example case the standard uncertainty of the contribution due to the interpolation of values for the
Transform Factor is taken as 0,29 dB.
The appropriate standard uncertainties from table 44 should be combined with uj30 by RSS in accordance with
TR 102 273 [3], part 1, sub-part 1, clause 5. This gives the combined standard uncertainty (uc EUT measurement) for the
EUT measurement in dB.
The value of uc EUT measurement is calculated as 3,56 dB.
ETSI
149 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.3.2.1 Expanded uncertainty for the receiver sensitivity measurement

The combined standard uncertainty of the results of the receiver sensitivity measurement is the combination of
uc Stripline attenuation measurement and uc EUT measurement.
uc receiver sensitivity measurement = 3,512 + 3,722 = 5,11 dB
The expanded uncertainty is ±1,96 x 5,11= ±10,0 dB at a 95 % confidence level.
6.9.3.3 Uncertainty contributions: Stage 2: Field measurement using a monopole

The second stage involves replacing the EUT with a field measuring device (either a monopole antenna or a 3-axis
probe) and, setting a particular output level from the signal generator (minimum or average), measuring the
corresponding field strength, see figure 39.
NOTE: In this case monopole field measurement involves mounting the monopole through a hole in the lower
plate of the Stripline (so that the feed point to the monopole is flush with the surface of the lower plate)
and measuring the field strength.
Central axis of
Stripline
Monopole
Signal
Load
generator
Ferrite beads 150 Ω

10dB termination
attenuators Receiving
device
Figure 39: Stage 2: Field measurement using a monopole

Mismatch in the transmitting and receiving parts: Whereas figure 39 shows schematically the equipment set-up for
field measurement using a monopole, figure 40 provides a detailed picture of the individual uncertainty contributions.
Mismatch: transmitting part: The mismatch uncertainty between the signal generator, signal generator cable, signal
generator attenuator and the Stripline input can be calculated from the approach described in annex G. The mismatch
strength measurement is subsequently performed since there are no changes from stage 1 to this part of the test set-up.
The standard uncertainty of the contribution due to the mismatch in the transmitting part is designated throughout all
NOTE: In this example case the standard uncertainty of the contribution due to mismatch in the transmitting part
The mismatch uncertainty in the receiving part i.e. between the monopole, receiving device attenuator, receiving device
cable and the receiving device is calculated from the approach described in annex G. This mismatch uncertainty
contributes only during the field measurement part of the test and therefore contributes to the combined standard
uncertainty.
ETSI
150 ETSI TR 100 028-1 V1.4.1 (2001-12)
Stripline
Stripline
Radiated
to monopole Monopole
Loss
coupling
Attenuator 1 Site Attenuator 2

10 dB effects 10 dB
Room
resonances, etc.
Cable Cable
+
+ factor
factor
Signal Cable
Receiving
Cable
generator loss loss device
cable 1 cable 2
Receiving
Signal Signal Receiving device
generator device
generator
Figure 40: Schematic of the field measurement using a monopole
Monopole: Input reflection coefficient: |ρM| = 0,333
Receiving device: Output reflection coefficient: |ρRD| = 0,20
All these contributions are U-distributed. Other contributions are (see annex G):
0,333 × 0,05 × 100

u j mismatch:antennaand attenuator = % = 1,177 %
2
ETSI
151 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,333 × 0,07 × 0,316 2 × 100

u j mismatch:antenna and cable2 = % = 0,165 %
2
0,333 × 0,2 × 0,316 2 × 0,8912 × 100

u j mismatch:antenna and receiving device = % = 0,373 %
2
u j mismatch:receiving part = 1,177 2 + 0,165 2 + 0,3732 = 1,25 %
The standard uncertainty of the contribution, due to the mismatch in the receiving part, is designated throughout all
level is designated throughout all parts of TR 102 273 [3] as uj38. Its value can be obtained from the manufacturer's data
sheet.
uncertainty from the manufacturer's data sheet is ±1,0 dB. As nothing is said about the distribution of this
TR 102 273 [3] as uj41.
the signal generator cable is taken as 0,00 dB since the uncertainty is systematic i.e. it is assumed constant
and common to both stages of the measurement.
ETSI
152 ETSI TR 100 028-1 V1.4.1 (2001-12)
uncertainty of the uncertainty due to cable factor, associated with each cable, is 0,5 dB provided that the precautions
detailed in the methods have been observed. i.e. routing and dressing of cables with ferrites. If no prevention has been
attempted the standard uncertainty is 4,0 dB (justification for these values is given in TR 102 273 [3], part 1,
sub-part 2, annex E). The standard uncertainty of the contribution due to the cable factor of the signal generator cable is
generator cable is taken as 0,00 dB since the precautions detailed in the methods have been observed and
the field strength is to be measured.
the contribution due to the insertion loss uncertainty of the signal generator attenuator is designated throughout all parts
of TR 102 273 [3] as uj40.
the signal generator attenuator is 0,00 dB since the precautions detailed in the methods have been
observed and the field strength is to be measured.
measurement frequency. The standard uncertainty of the contribution due to the ambient effect is designated throughout
all parts of TR 102 273 [3] as uj34. The value of the standard uncertainty should be taken from table 45.

(EUT OFF) is within: the contribution
the recommendations given in EN 55020 [7] regarding positioning of the Stripline and layout of the absorber. These can
taken to be 0,00 dB since in this test method, site effects contribute equally to both stages as the
recommendations are assumed to have been fully observed.
ETSI
153 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.3.3.2.5 Antenna factor of the monopole
Stripline: antenna factor of the monopole: This has been derived from measurements taken within the Stripline.
Therefore, the given values incorporate several of the field disturbance factors which the Stripline possesses and which
therefore do not have to be allowed for as individual contributions. Amongst these included effects are imaging,
characteristic impedance of the line, non-planar nature of the field etc. The standard uncertainty of the contribution due
to the antenna factor of the monopole is designated throughout all parts of TR 102 273 [3] as uj30.
NOTE: In this example case the standard uncertainty of the contribution due to the antenna factor of the
monopole is taken as 1,15 dB. This combined uncertainty source is only present in the actual
measurement.
6.9.3.3.2.6 Monopole attenuator
Insertion loss: Monopole attenuator: The monopole attenuator has an insertion loss as well as an uncertainty
the attenuator is not used outside the manufacturer's specification. The standard uncertainty of the contribution due to
the insertion loss uncertainty of the monopole attenuator is designated throughout all parts of TR 102 273 [3] as uj40.
the monopole attenuator is taken as 0,1 dB since the uncertainty contributes only to stage 2 of this test
method.
contribution due to the insertion loss uncertainty of the receiving device cable is designated throughout all parts of
TR 102 273 [3] as uj41.
the receiving device cable is taken as 0,15 dB since the uncertainty is only present in the second stage of
this test method.
influence on the measuring system including its interaction with the site. It consists of the leakage caused by cable
screening inefficiency and introducing an unbalanced current into the monopole. In a radiated measurement the
standard uncertainty of the cable factor, associated with each cable, is 0,5 dB provided that the precautions detailed in
the methods have been observed. i.e. routing and dressing of cables with ferrites. If no prevention has been attempted
the standard uncertainty is 4,0 dB (justification for these values is given in TR 102 273 [3], part 1, sub-part 2,
annex E). The standard uncertainty of the contribution due to the cable factor of the receiving device cable is designated
observed.
of TR 102 273 [3] as uj47.
NOTE 1: In this example case the standard uncertainty of the contribution due to the receiving device absolute level
uncertainty (a range change is assumed) is obtained from the manufacturers data as ±0,5 dB. As nothing
is said about the distribution of this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1,
sub-part 1, clause 5.1.2) in logs is assumed, and the standard uncertainty is calculated as 0,29 dB.
ETSI
154 ETSI TR 100 028-1 V1.4.1 (2001-12)
taken as 0,00 dB.

Random uncertainty: The magnitude can be assessed from multiple measurements of the field strength measurement.
The standard uncertainty of the contribution due to the random uncertainty is designated throughout all parts of
TR 102 273 [3] as ui01.
The field strength measurement was repeated 10 times. The following results were obtained in dBµV (before correcting
for cabling and attenuator network insertion loss):
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3;
Y−
X2
32,10 × 10 −6
−
(17,77 ×10 )
−3 2

n −1 10 − 1
is:
238,3 × 10 −6 100
−3
1,777 × 10 11,5
ETSI
155 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.9.3.3.4 Summary table of contributions

Table 46: Contributions from the monopole field measurement

uj31 Stripline: antenna factor of the monopole 1,15
uj40 insertion loss: monopole attenuator 0,10
ui01 random uncertainty 1,17
sub-part 1, clause 5. This gives the combined standard uncertainty (uc field measurement using a monopole) for the field
measurement using a monopole in dB.
The value of uc field measurement using a monopole is calculated as 1,91 dB.

The combined standard uncertainty of the results of the receiver sensitivity measurement is the combination of the
components outlined in clauses 6.9.3.1.4 and 6.9.2.4.4. The components to be combined are uc measurement of the EUT and
uc field measurement using a monopole.
u c = 2,18 2 + 1,912 = 2,90 dB
6.9.3.4 Uncertainty contributions: Stage 2: Field measurement using 3-axis probe

In this case, field measurement involves the use of a 3-axis probe and measuring the vertical component of the electric
field.
ETSI
156 ETSI TR 100 028-1 V1.4.1 (2001-12)

Whereas figure 41 shows schematically the equipment set-up for field measurement using a 3-axis probe, figure 42
provides a detailed picture of the individual uncertainty contributions.
Central axes of
Stripline
3 axis probe
Signal
Load
generator
Ferrite beads
10dB
Digital 150 Ω
attenuator termination
readout
Figure 41: Stage 2: Field measurement using a 3-axis probe
Mismatch: transmitting part: The mismatch uncertainty between the signal generator, signal generator cable, signal
generator attenuator and the Stripline input can be calculated from the approach described in annex G. The mismatch
strength measurement is subsequently performed since there are no changes from stage 1 to this part of the test set-up.
The standard uncertainty of the contribution due to the mismatch in the transmitting part is designated throughout all
NOTE 1: In this example case the standard uncertainty of the contribution due to mismatch in the transmitting part
ETSI
157 ETSI TR 100 028-1 V1.4.1 (2001-12)
Stripline
Stripline
Radiated
to 3 axis probe 3 axis probe
Loss coupling
Mismatch +
Digital
Attenuator 1
Site readout
10 dB effects
Room Indicating
resonances, etc. device
Mismatch +
Cable +
factor
Signal Cable
generator loss
cable 1
Mismatch +
Signal Signal
generator
generator
Figure 42: Schematic of the field measurement using a 3-axis probe
level uncertainty is designated throughout all parts of TR 102 273 [3] as uj38. Its value can be obtained from the
manufacturer's data.
NOTE 1: In this example case the uncertainty of the contribution due to the signal generator absolute output level is
ETSI
158 ETSI TR 100 028-1 V1.4.1 (2001-12)
TR 102 273 [3] as uj41.
the signal generator cable is taken as 0,00 dB as a field strength measurement is performed using the
3-axis probe.
uncertainty of the uncertainty due to cable factor, associated with each cable, is 0,5 dB provided that the precautions
detailed in the methods have been observed. i.e. routing and dressing of cables with ferrites. If no prevention has been
attempted the standard uncertainty is 4,0 dB (justification for these values is given in TR 102 273 [3], part 1,
sub-part 2, annex E). The standard uncertainty of the contribution due to the cable factor of the signal generator cable is
generator cable is taken as 0,00 dB since the precautions detailed in the methods have been observed and
the field strength is to be measured.
the contribution due to the insertion loss uncertainty of the signal generator attenuator is designated throughout all parts
of TR 102 273 [3] as uj40.
the signal generator attenuator is 0,00 dB since the precautions detailed in the methods have been
observed and the field strength is to be measured.
ETSI
159 ETSI TR 100 028-1 V1.4.1 (2001-12)
TR 102 273 [3] as uj34. The values of the standard uncertainties should be taken from table 47.

the recommendations given in EN 55020 [7] regarding positioning of the Stripline and layout of absorber. These can
taken to be 0,00 dB since in this test method, site effects contribute equally to both stages as the
recommendations are assumed to have been fully observed.
Stripline: characteristic impedance: This uncertainty contribution results from the difference between the free-space
wave impedance (377 Ω) for which the 3-axis probe had been developed and that for the Stripline (150 Ω). The
standard uncertainty of the contribution due to the characteristic impedance of the Stripline is designated throughout all
NOTE 3: In this example case the standard uncertainty of the contribution due to the characteristic impedance of
the Stripline is taken as having a standard uncertainty of 0,58 dB.
Stripline: mutual coupling of the 3-axis probe to its image in the plates: This contribution is to take account of the
fact that the probe has images in both plates of the Stripline. The standard uncertainty of the contribution due to the
mutual coupling of the 3-axis probe to its image in the plates is designated throughout all parts of TR 102 273 [3] as
uj25.
NOTE 4: In this example case the uncertainty of the contribution due to the mutual coupling of the 3-axis probe to
its image in the plates is taken as ±0,5 dB. As nothing is said about the distribution of this uncertainty, a
rectangular distribution (see TR 102 273 [3], part 1, sub-part 1, clause 5.1.2) in logs is assumed, and the
standard uncertainty is calculated as 0,29 dB.
6.9.3.4.2.5 3-axis probe field measurement
Stripline: field strength measurement as determined by the 3-axis probe: The standard uncertainty of the
contribution, due to the field strength measurement uncertainty as determined by the 3-axis probe, is designated
throughout all parts of TR 102 273 [3] as uj28. Its value can be derived from the manufacturers data sheet.
NOTE: In this example case the uncertainty of the contribution due to the field strength measurement as
determined by the 3-axis probe is obtained from the manufacturer's data sheet as ±1 dB. As nothing is
said about the distribution of this uncertainty, a rectangular distribution (see TR 102 273 [3], part 1,
sub-part 1, clause 5.1.2) in logs is assumed, and the standard uncertainty is calculated as 0,58 dB.
ETSI
160 ETSI TR 100 028-1 V1.4.1 (2001-12)

Random uncertainty: The magnitude can be assessed from multiple measurements of the receiver sensitivity. The
TR 102 273 [3] as ui01.
The field strength measurement was repeated 10 times. The following results were obtained in dBµV (before correcting
for cabling and attenuator network insertion loss):
- 65,4; 63,4; 66,0; 65,3; 63,0; 64,9; 65,2; 66,8; 65,5; 63,7.
- 1,862 × 10-3; 1,479 × 10-3; 1,995 × 10-3; 1,841 × 10-3; 1,413 × 10-3; 1,758 × 10-3; 1,820 × 10-3; 2,188 × 10-3;
1,884 × 10-3; 1,531 × 10-3;
Y−
X2
32,10 × 10−6 −
(17,77 ×10 )
−3 2

n −1 10 − 1
is:
238,3 ×10 −6 100

−3
1,777 ×10 11,5
evaluated as 1,17 dB.
See also the note relating to the random uncertainty in clause 6.4.7.

Table 48: Contributions from the 3-axis probe field measurement

uj26 Stripline: characteristic impedance 0,58
uj25 Stripline: mutual coupling of the 3-axis probe to its image in the plates 0,29
uj28 Stripline: field strength measurement as determined by the 3-axis probe 0,58
ETSI
161 ETSI TR 100 028-1 V1.4.1 (2001-12)
sub-part 1, clause 5. This gives the combined standard uncertainty (uc 3-axis probe) for the receiver sensitivity
measurement in dB.
The value of uc 3-axis probe is calculated as 1,65 dB.

The combined standard uncertainty of the results of the receiver sensitivity measurement is the combination of the
components outlined in clauses 6.9.3.1.4 and 6.9.3.4.4. The components to be combined are uc EUT measurement and
uc 3-axis probe.
uc = 2,182 + 1,652 = 2,73 dB
6.10 Uncertainty of fully automated test systems.

So far the uncertainty calculations for manual measurements have been examined.
But in many radio technologies testing is performed using fully automated test systems. In technologies such as GSM,
DECT and Bluetooth, certification and type approval is based on measurements using such test systems. This gives an
improvement in reproducibility and test time compared to manual measurements, but the measurement uncertainty for
such test systems has yet to be documented.
One major reason is that the procedures and calculations outlined for the more simple test methods do not cover fully
automated test systems due to the complexity, even though the basic principles still apply.
The measurements are basically carried out in the same way as the manual measurements. A conducted power
measurement is still performed by connecting the EUT to a power measuring instrument through a combining network
consisting of cables, attenuators and maybe filters. Then a power measurement is carried out, and a correction factor is
applied to the reading of the instrument to get the final test result.
Similarly a receiver measurement is done by connecting one or more RF signal generators to an EUT through a
combining network and adjusting the output levels from the generators each time using correction factors.
The major difference between the manual measurement and the fully automated test system measurement is how this
correction factor is derived. For fully automated systems this is normally done by executing Path Compensation
procedures.
The purpose of path compensation procedures is (as mentioned) to generate correction factors, and in a well designed
test system these correction factors eliminate all errors leaving "only" some irreducible stability and mismatch errors.
In most fully automated test systems the path compensation procedures are a combination of measurements performed
at the same time as the actual measurement as well as periodic measurements on sub parts of the test system, but to have
a full picture of the uncertainties involved the path compensation and the actual measurement should be seen as one
procedure. With a well designed overall path compensation procedure it is easy to see that if all instruments and
components were stable and linear and had an impedance of exactly 50 Ohms then the only uncertainty contribution
would be the absolute uncertainty of the power meter.
But as with the manual measurements the instruments are not totally stable, and there are mismatch uncertainties due to
non-ideal coaxial components.
ETSI
162 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.10.1 Test system properties

A fully automated test system normally consists of a set of test instruments (usually exactly the same as the ones used in
the manual measurements), but in addition it contains a switch unit. The purpose of the switch unit is to create the
correct set-ups using attenuators, power combiners, filters, amplifiers, and cables. The set-ups inside the switch unit is
then realized using RF switches controlled by a system controller – normally a PC with appropriate test software. A
switch unit often consists of more than 100 components.
A fully automated test system can perform all the common radio tests: Transmitter tests including output power, timing,
modulation, output spectrum, and spurious emissions, and receiver tests including 1, 2, and 3 signal measurements as
the test system contains 3 RF signal generators. The BER measurement or the signal-to-noise measurement is not
shown. It is assumed to be either a part of signal generator 1 or some external equipment connected to the base band
output of the EUT. It is not important for the analysis of the test system because once the level uncertainties are
calculated, the rest (BER , signal-to-noise, modulation, or timing) are the same as with the manual measurements.
Signal generator 1 generates the wanted signal
Signal generator 2 generates the low frequency unwanted signal
Signal generator 3 generates the high frequency unwanted signal
(Both signal generator 2 and 3 can also produce in-band signals for 3 signal measurements.)
The signal analyser is capable of measuring both power, frequency, and modulation, but for the purpose of this analysis
only power is considered.
Tests are normally carried out as follows: The EUT is connected to a specific EUT connector on the test system. Then
the test operator selects and activates some tests, and some test results are produced by the test system. Depending on
the degree of automation the operator may be prompted to control the EUT from time to time – for instance to set up a
connection with the EUT or to switch the EUT to a different channel.
It is normally not visible to the operator how the tests are carried out by the test system, but this is often described in the
test system documentation. As a part of a test the path compensation related to that test can be run prior to the actual
testing. This depends on the flexibility and complexity of the switch unit and the test software.
6.10.2 General aspects of the measurement uncertainty

As indicated previously, the main difference between manual testing and a fully automated test system is how the
correction factors are generated. From a measurement uncertainty point of view this is very important because this is X
one of the major contributions to the overall RF level uncertainty of the actual measurement. The other contributions
are:
- Instruments stability;
- Instruments linearity;
- Mismatch between the EUT and the test system;
- RF switch repeatability.
Often it is the power meter that is the essential instrument in the path compensation, and the one which provides the
traceability to external standards.
The uncertainty of the correction factor is very dependant on how the correction factor is measured. The contributions
to the uncertainty are:
- Absolute power meter uncertainty;
- Instruments stability;
- Instruments linearity;
- Mismatch between the instruments and the individual components of the switch unit;
- Errors due to interpolation between correction factors at different frequencies.
ETSI
163 ETSI TR 100 028-1 V1.4.1 (2001-12)
As will be shown, the mismatch uncertainty is the most complex component of the overall measurement uncertainty. It
can for complex fully automated systems be the combination of several thousands of individual mismatch contributions.
The amount of contributions can, however, be reduced by disregarding very small contributions.
The contributions tend to be greater than with the manual measurements because more cables and switches are
necessary to provide the needed flexibility.
For the purpose of the measurement uncertainty analysis, two fully automated test systems will be considered: a
"simple" test system, and a "complex" test system.
6.10.3 The "simple" test system

The test system is shown in figure 43:
Filters and attenuators

Signal analyser
Switch 2 EUT conn. EUT
Generator 1
Power combiner Switch 1

Generator 2
Sig. gen. out
Generator 3
Figure 43: The "simple" test system
6.10.3.1 Transmitter measurement

For the "simple" test system the path compensation procedures and the actual measurement for transmitter
measurements using the correction factors, are as follows:
The path compensation is performed as follows:
- Switch 1 is set so the generators are connected to the Sig. gen. out connector.
- Switch 2 is set so the signal analyser is connected to the EUT connector.
Measurement 1:
1) A power meter is connected to the RF out connector through a cable and a 10 dB attenuator.
2) The RF generators are in turn adjusted to a suitable level which gives a reading in the operational range of the
power meter. When one generator is active the others are turned down, so they do not contribute to the
measurement, but their output impedance is still 50 ohms. A series of measurements covering the frequency
range of interest is carried out and for each frequency the reading is stored by the test system.
ETSI
164 ETSI TR 100 028-1 V1.4.1 (2001-12)
Measurement 2:
1) The power meter is removed and the open end of the 10 dB attenuator is connected to the EUT connector. For all
the frequencies and generator level settings in step 4 the power level is measured by the signal analyser. This is
preferably done with the same analyser setting as the one used in the actual EUT measurement. The readings are
stored by the test system. For each frequency the correction factor is calculated as the difference (in dB) between
the signal analyser reading XX and the power meter XX.
Measurement 3: ( The actual measurement):
1) The EUT is connected to the EUT connector.
2) The power level of the signal generated by the EUT is measured, and the signal analyser reading is stored by the
test system.
3) The final result is then calculated as the reading from step 2 (in dBm) minus the correction factor calculated in
the path compensation procedure at the appropriate frequency. (If a correction factor at the measuring frequency
does not exist it is found by interpolation between the two correction factors at the nearest frequencies on each
side.)
6.10.3.1.1 Error analysis

The combined path compensation procedure and the actual test consists of 3 individual measurements as shown in
figure 44: two measurements in the path compensation part and one in the actual measurement.
In each of the 3 measurements a signal source is connected to a measuring instrument through a network consisting of
several components and a level is measured.
In the following the total procedure is analysed.
The following assumptions apply for the analysis:
- The generator has a static error of Egen dB in measurement 1 (compared to the setting of the generator level);
- Between measurement 1 and measurement 2 there is a generator drift error dEgen dB;
- The attenuation between the generator and the sig. gen. out connector is Att1 dB;
- Between measurement 1 and measurement 2 there is an attenuation change in the network between the generator
and the sig. gen. out. connector of dAtt1 dB;
- The attenuation of the external cable and attenuator is Att2 dB;
- Between measurement 1 and measurement 2 there is an attenuation change in the external cable and attenuator of
dAtt2 dB;
- There is a static error of Epow dB in power measurements using the power meter;
- The attenuation between the EUT connector and the signal analyser i Att3 dB;
- Between measurement 2 and measurement 3 there is an attenuation change in the network between the EUT
connector and the signal analyser of dAtt3 dB;
- There is a static error of Esa dB in the power measurement in measurement 2 using the signal analyser;
- Between measurement 2 and measurement 3 there is a signal analyser drift error dEsa dB;
- The EUT has an output power of Pout dBm;
- The generator level is set to Pgen dBm in measurement 1 and 2;
- If the value read from the signal analyser in measurement 3 differs from the value in measurement 2 there is a
signal analyser linearity (or log fidelity) error dElog.
In measurement 1 the reading on the power meter is:
P1 = Pgen + Egen – Att1 – Att2 + Epow dBm
ETSI
165 ETSI TR 100 028-1 V1.4.1 (2001-12)
In measurement 2 the reading on the signal analyser is:
P2 = Pgen + Egen + dEgen – Att1 – dAtt1 – Att2 – dAtt2 – Att3 + Esa dBm
The correction factor is:
Ccorr = P2 – P1 =
(Pgen + Egen + dEgen – Att1 – dAtt1 – Att2 – dAtt2 – Att3 + Esa) – (Pgen + Egen – Att1 – Att2 + Epow) =
dEgen – dAtt1 – dAtt2 – Att3 + Esa – Epow dB
In measurement 3 the reading from the signal analyser is:
P3 = Pout – (Att3 + dAtt3) + Esa + dEsa + dElog dBm
The measured result after having applied the correction factor to the reading from measurement 3 is:
Pmeas = P3 – Ccorr =
Pout – (Att3 + d Att3) + Esa + dEsa + dElog – (dEgen – dAtt1 – dAtt2 – Att3 + Esa – Epow) =
Pout + dAtt3 + dEsa +dElog – dEgen + dAtt1 + dAtt2 + Epow dBm
As can be seen from the calculated result all static errors in the combined measurement except the power meter error
have cancelled. Apart from that only the drift and linearity errors remain.
The remaining errors are:
- the absolute uncertainty of the power meter
- the linearity (or log fidelity) of the signal analyser due to the fact that the level measured by the signal analyser
in actual measurement may be different from the level measured in the path compensation
- signal analyser drift between the different measurements
- signal generator drift between the different measurements
- repeatability of the switches in the switch unit
- change of the insertion losses between the different measurements
Since the path compensation is performed at discrete frequencies there is an additional error
- error due to interpolation between correction factors at different frequencies
Finally, in addition to the uncertainties mentioned there is a mismatch uncertainty in each measurement.
The mismatch uncertainty is analysed in clause 6.10.3.1.2
6.10.3.1.2 Mismatch uncertainty

For each measurement there is a mismatch uncertainty which is the combination of all the mismatch uncertainties
between all of the parts in the path between the signal source and the measuring instrument.
Fortunately many of the mismatch uncertainties are cancelled due to the total procedure.
Firstly the two measurements involved in the path compensation procedure are considered. The correction factor is the
difference between the two values measured; this means that the total error is the difference between the errors in the
two measurements, so all errors which are identical cancel.
ETSI
166 ETSI TR 100 028-1 V1.4.1 (2001-12)
Switch unit
Signal Attenuators,
analyser cables, and
Switch 2 EUT conn.
filters
Generator Combiner and Switch 1 Sig. gen. out

cables Power
10 dB att
meter
Measurement 1:Path compensation: power meter reading
Switch unit
Signal Attenuators, EUT conn.
analyser cables, and
Switch 2 10 dB att
filters

cables
Measurement 2:Path compensation: signal analyser reading
Switch unit
Signal Attenuators,
analyser cables, and EUT conn.
filters Switch 2 EUT

cables
Measurement 3: The actual measurement
Figure 44: The three set-ups in the transmitter measurement
From figure 44 it can be seen that all the mismatch uncertainties from the path between the 10 dB attenuator and the RF
signal generators cancel because they are present in both measurements 1 and 2.
In measurement 1 all the mismatch uncertainties associated with the power meter remain. The rest are cancelled.
In measurement 2 all mismatch uncertainties from the path between the 10 dB attenuator and the signal analyser remain,
as they only appear here. For the same reason all the mismatch uncertainties where one of the parts is to the right of the
EUT connector, and the other part is to the left of the EUT connector remain.
Then when measurement 3 (the actual measurement) is taken into account it can be seen that parts of the mismatch
uncertainties from measurement 2 cancel, since they are also present in measurement 3: all the uncertainties from the
path between the EUT connector and the signal analyser.
What is left in measurement 3 are all the mismatch uncertainties where the EUT is one of the parts. The rest cancel with
measurement 2.
ETSI
167 ETSI TR 100 028-1 V1.4.1 (2001-12)
The total mismatch uncertainty in the total measurement including the path compensation is then the combination of the
following part uncertainties:
1) all mismatch uncertainties where the power sensor is one of the parts (measurement 1);
2) all mismatch uncertainties where the two parts are on each side of the EUT conn. (measurement 2);
3) all mismatch uncertainties where the EUT is one of the parts (measurement 3).
Based on this the calculation of the total mismatch uncertainty can be done in two ways.
If all VSWRs and insertion losses (or gains) of the individual components in the test system are known all the
contributions can be calculated and combined.
But a more simple approach is to measure (or estimate by other methods) the reflection coefficient Rg of the free end of
the 10 dB attenuator, measure (or take from the specification sheet) the reflection coefficient Rp of the power meter,
measure the reflection coefficient Ri of the EUT connector, and measure (or estimate) the reflection coefficient Reut of
the EUT.
If these 4 reflection coefficients are known, the total uncertainty is the combination of
- Ri*Rp/√2 (from measurement 1);
- Rg*Ri//√2 (from measurement 2);
- Reut*Ri//√2 (from measurement 3).
This is exactly the same result as if the measurement had been done manually with a generator, a power meter, and a
signal analyser if the switch unit paths are considered as parts of the individual instruments.
But the method of analysing the "simple" test system is important, because the same method is used in the "complex"
test system, and here the result are not similar to any simple manual measurement.
6.10.3.2 Receiver measurements

For the "simple" test system the path compensation procedures for receiver measurements and the actual measurement
using the correction factors, are as follows:
The path compensation (measurement 1) is done as follows:
1) Switch 1 and switch 2 is set so the generators are connected to the EUT connector.
2) Then the power meter is connected to the EUT connector.
3) The RF generators are in turn adjusted to a suitable level which gives a reading in the operational range of the
power meter. When one generator is active the others are turned down, so they do not contribute to the
measurement, but their output impedance is still 50 ohms. A series of measurements covering the frequency
range of interest is done and for each frequency the reading are stored by the test system.
4) For each frequency point the correction factor is calculated as the difference (in dB) between the power meter
reading and the generator setting.
The actual measurement (measurement 2) is done as follows:
2) The generator is set to the wanted signal level (in dBm) minus the correction factor at the appropriate frequency.
(If a correction factor at the measuring frequency does not exist it is found by interpolation between the two
correction factors at the nearest frequencies on each side.)
3) Then the appropriate receiver measurement is done (BER or signal-to-noise ratio).
ETSI
168 ETSI TR 100 028-1 V1.4.1 (2001-12)

figure 45: two measurements in the path compensation part and one in the actual measurement.
- The generator has an static error of Egen dB in measurement 1(compared to the setting of the generator level).
- The generator has a linearity/log fidelity error of Elog dB between the levels in measurement 1 and measurement
2.
- Between measurement 1 and measurement 2 there is a generator drift error dEgen dB.
- The attenuation between the generator and the EUT connector is Att1 dB in measurement 1.
- Between measurement 1 and measurement 2 there is an attenuation change in the network between the generator
and the EUT connector of dAtt1 dB.
- There is a static error of Epow dB in the power meter measurement in measurement 1.
- The generator level is set to Pgen1 dBm in measurement 1.
- The wanted level at the EUT connector is Pwanted dBm in the actual measurement.
In measurement 1 the reading on the power meter is:
P1 = Pgen1 + Egen – Att1 + Epow dBm
The correction factor is calculated to be Ccorr = Pgen1 – P1 =
Pgen1 - (Pgen1 + Egen – Att1 + Epow) = -Egen + Att1 – Epow
In measurement 2 the generator level is set to Pgen2 = Pwanted + Ccorr =
Pwanted -Egen + Att1 – Epow dBm
The level at the EUT connector in the actual measurement is
Peut = Pgen2 + Egen + dEgen + Elog – Att1 – dAtt1 =
Pwanted -Egen + Att1 – Epow + Egen + dEgen + Elog – Att1 – dAtt1 =
Pwanted – Epow + dEgen + Elog – dAtt1
As can be seen from the calculated result, again all static errors in the combined measurement except the power meter
error have cancelled. Apart from that only the drift and linearity errors remain.
- the absolute uncertainty of the power meter
- the linearity (or log fidelity) of the signal generator due to the fact that the level setting of the generator in actual
measurement may be different from the level setting in the path compensation
- change of the insertion loss between the path compensation and the actual measurement including repeatability
of the switches in the switch unit
Since the path compensation is performed at discrete frequencies there is an additional error:
ETSI
169 ETSI TR 100 028-1 V1.4.1 (2001-12)
The mismatch uncertainty is analysed in clause 6.10.3.2.2
6.10.3.2.2 Mismatch uncertainty
Switch unit
Signal Attenuators,
Power
Switch 2
filters meter

cables
Path compensation: power meter reading
Switch unit
Signal Attenuators,
Switch 2 EUT
filters

cables
The actual measurement
Figure 45: The two set-ups in a receiver measurement
As can be seen from figure 45, performing a similar analysis as with the transmitter measurement, the remaining
mismatch uncertainty from the first measurement is the combination of all the mismatch uncertainties associated with
the power meter, and from the actual measurement all the mismatch uncertainties associated with the EUT.
Again, based on this the calculation of the total mismatch uncertainty can be done in two ways.
If all VSWRs and insertion losses (or gains) of the individual components in the test system are known all the
contributions can be calculated and combined.
But a more simple approach is to measure (or estimate by other methods) the reflection coefficient Ro of the EUT
connector, measure (or take from the specification sheet) the reflection coefficient Rp of the power meter, and measure
(or estimate) the reflection coefficient Reut of the EUT.
If these 3 reflection coefficients are known, the total uncertainty is the combination of
- Ro x Rp/√2 (from the path compensation)
- Reut x Rp//√2 (from the actual measurement)
ETSI
170 ETSI TR 100 028-1 V1.4.1 (2001-12)
Again exactly the same result as if the measurement had been done manually with a generator and power meter if the
switch unit paths are considered as parts of the individual instruments.
For the "simple" test system it was not necessary to go through this lengthy analysis to get the uncertainty, because the
analogy to the simple measurements can be directly seen. But the method is important to understand and to use, because
it is needed for the analysis of more complex test systems where the similarity to simple measuring set-ups does not
exist.
6.10.4 The "complex" test system

As the "simple" test system, the "complex" system consists of a set of measuring instruments and a switch unit.
And as with the "simple" test system the RF level traceability is provided by very accurate power meters rather than the
other RF instruments.
The main difference between the "simple" and the "complex" test system is that the path compensation procedures and
RF level setting procedures are more complex and involve reference and switch points inside the switch unit, which
cannot be accessed from the outside.
The benefit is that most of the path compensation procedures can be done directly in connection with the actual
measurements without need of test operator intervention. This reduces the potential stability errors which can be present
in measurements with "simple" test systems due to the time between path compensation and measurements.
The error sources are generally the same for the two types of test systems, and the methods used to perform the analysis
are the same, but the mismatch uncertainty is very complex, and can be difficult to estimate.
Power
sensor
A
1 Ext. signalling device

1 Att. 1
2 2 10 dB
50 Ohm S3
3 3
S26 S25 S8
Signal 4 4 6 dB Dir. coupler 1 EUT conn.
comb.
analyser 5 5 1 Dir. coupler 2
S1 S2 S4 S5
6 6
50 Ohm
S24 S23 S17 S7
S22 S21 S20 S19
S16
S6
Gen 1 50 Ohm
S18
1 1
Att. 3 Att. 4 Att. 2
20 dB 20 dB 2 2 10 dB
S27 3 3 S28
S11
4 4
Gen 2 5 5
6 6
6 dB
Power comb.
splitter 2 Sig.gen.out conn.
50 Ohm
S10 S9
Gen 3
S12
Figure 46: A complex test system
Figure 46 shows a "complex" type of test system which is capable of doing all normal RF measurements. More
complicated test systems exist, but the following analysis will be similar for all of them.
ETSI
171 ETSI TR 100 028-1 V1.4.1 (2001-12)
The path compensation for this test system consists of two procedures: an external path compensation procedure which
requires the test operator to connect cables and power meter to some external connectors and an internal path
compensation procedure. This external path compensation characterizes a small part of the switch unit consisting of
only cables, attenuators, and switches – in other words passive components which can be assumed to be stable over a
relatively long time period. For transmitter measurements this part of the switch unit is the part from switch S5 to the
EUT connector.
For receiver measurements it is the part between switch S4 and the EUT connector.
The rest of the switch unit and the instruments are covered by internal path compensation procedures which do not
require test operators intervention, and they are run prior to the actual measurements as an integral part of each test. As
with the "simple" test system the traceability is provided by an external power meter, which is the only instrument
where the absolute uncertainty is important. Any systematic errors in the other instruments are compensated for.
The error sources are basically the same as with the "simple" test system:
- Absolute power meter uncertainty.
- Instruments stability.
- Instruments linearity.
- Mismatch between the instruments and the individual components of the switch unit.
- Errors due to interpolation between correction factors at different frequencies.
The difference compared to a "simple” test system is that the mismatch uncertainty is more complex because there are
more procedures involved in the testing and path compensations and because some of the reflection coefficients of
interest are inside the switch unit. To measure them, the switch unit would have to be disassembled.
6.10.4.1 Receiver measurements

For the purpose of analysing receiver measurements the "complex" test system can be simplified as shown in figure 47:
Power
meter A
Sub network C
Generator Sub network A EUT conn EUT

Sub network B
S4
Switch unit
Figure 47: Model for analysis of receiver measurements
In figure 47:
Sub network A consists of everything between a generator and switch S4.
Sub network B consists of everything between switch S4 and the EUT connector.
Sub network C consists of everything between switch S4 and power meter A.
ETSI
172 ETSI TR 100 028-1 V1.4.1 (2001-12)
Each sub network contain cables, switches, attenuators, filters, and other components.
The external path compensation is performed as follows. This is not done in connection with every measurement, but
may be done with for instance 3 month intervals.
Measurement 1:
1) A power meter (power meter B) is connected to the EUT connector. Switch S4 is set so the generator is
connected to power meter B.
2) The RF generator is set to a level which gives a reading in the operational range of the power meter. When one
generator is active the others are turned down, so they do not contribute to the measurement, but their output
impedance is still 50 ohms. A series of measurements covering the frequency range of interest is done and for
each frequency the reading are stored by the test system.
Measurement 2:
1) Then switch S4 is set so the generator is connected to power meter A.
2) For all the frequencies and generator level settings in measurement 1 step 2 the power level is measured by
power meter A. The readings are stored by the test system
3) For each frequency the correction factor is calculated as the difference (in dB) between the reading from power
meter B and power meter A. These are the external correction factors (path compensation data) stored by the test
system.
The internal path compensation is performed as follows. This is done immediately prior to every measurement as an
integral part of the test case.
Measurement 3:
1) Switch S4 is set so the generator is connected to power meter A.
2) Then the RF generator is set to a level which gives a reading in the operational range of the power meter. When
one generator is active the others are turned down, so they do not contribute to the measurement, but their output
impedance is still 50 ohms. A series of power meter readings and generator level settings covering the frequency
range of interest is done and for each frequency the reading and setting are stored by the test system.
3) For each frequency the correction factor is calculated as the difference (in dB) between the power meter reading
and the generator level setting. These are the internal correction factors (path compensation data) stored by the
test system.
The actual test is performed as follows. (For 2 or 3 signal measurements the following level setting procedure is done
for each signal generator).
Measurement 4:
2) Switch S4 is set so the generator is connected to the EUT connector.
3) The generator is set to the wanted signal level (in dBm) minus the external and the internal correction factor at
the appropriate frequency. (If a correction factor at the measuring frequency does not exist it is found by
interpolation between the two correction factors at the nearest frequencies on each side.)
4) Then the appropriate receiver measurement is done (BER or signal-to-noise ratio).
ETSI
173 ETSI TR 100 028-1 V1.4.1 (2001-12)

figure 48 to figure 50: two measurements in the external the path compensation part, one in the internal path
compensation and one in the actual measurement.
- The generator has an static error of Egen1 dB in the measurement 1 (compared to the setting of the generator
level).
- Between measurement 1 and measurement 2 there is a generator drift error dEgen1 dB.
- The generator has an static error of Egen2 dB in the measurement 3(compared to the setting of the generator
level).
- Between measurement 3 and measurement 4 there is a generator drift and linearity error dEgen2 dB (The
generator level may not be the same in the path compensation and the measurement – therefore the linearity/log
fidelity error).
- The attenuation between the generator and the switch S4 is AttD dB in measurement 1.
- Between measurement 1 and measurement 2 there is an attenuation change in AttD of dAttD dB.
- The attenuation between the input of switch S4 and the EUT connector is AttB dB in measurement 1.
- Between measurement 1 and measurement 4 there is an attenuation change in AttB of dAttB dB.
- The attenuation between the input of switch S4 and power meter A is AttC dB in measurement 2.
- Between measurement 2 and measurement 3 there is an attenuation change in AttC of dAttC dB.
- The attenuation between the generator and the switch S4 is AttA dB in measurement 3.
- Between measurement 3 and measurement 4 there is an attenuation change in AttA of dAttA dB.
- There is a static error of EpowA dB in the power meter A measurement in measurement 2.
- Between measurement 2 and measurement 3 there is a change in EpowA of dEpowA.
- The generator level is set to Pgen1 dBm in measurement 1 and 2.
- The wanted level at the EUT connector is Pwanted dBm in the actual measurement.
In measurement 1 the reading from power meter B is:
P1 = Pgen1 + Egen1 – AttD – AttB + EpowB
In measurement 2 the reading from power meter A is:
P2 = Pgen1 + Egen1 + dEgen1 – AttD – dAttD – AttC + EpowA
The external correction factor
Ccorr1 = P1 – P2 =
(Pgen1 + Egen1 – AttD – AttB + EpowB) - (Pgen1 + Egen1 + dEgen1 – AttD – dAttD – AttC + EpowA) =
- dEgen1 + dAttD + AttC – EpowA – AttB + EpowB
ETSI
174 ETSI TR 100 028-1 V1.4.1 (2001-12)
P3 = Pgen2 + Egen2-AttA-AttC-dAttC + EpowA + dEpowA
The internal correction factor
Ccorr2 = P3 – Pgen2 = Pgen2 + Egen2-AttA-AttC-dAttC + EpowA + dEpowA – Pgen2 =
Egen2-AttA-AttC-dAttC + EpowA + dEpowA
In measurement 4 (the actual measurement) the generator level is set to
Pgen3 = Pwanted – Ccorr1 – Ccorr2
The level at the EUT connector is:
P4 = Pgen3 + Egen2 + dEgen2 – AttA – dAttA – AttC – dAttC =
Pwanted – Ccorr1 – Ccorr2 + Egen2 + dEgen2 – AttA – dAttA – AttB – dAttB =
Pwanted – (-dEgen1 + dAttD + AttC-EpowA-AttB + EpowB) – (Egen2-AttA-AttC
-dAttC + EpowA + dEpowA) + Egen2 + dEgen2 – AttA – dAttA – AttB – dAttB =
Pwanted + dEgen1 – dAttD - EpowB + dAttC - dEpowA + dEgen2 – dAttA – dAttB
Again, as can be seen from the calculated result, again static errors in the combined measurement except the power
meter B error have cancelled. Apart from that only the drift and linearity errors remain.
- the absolute uncertainty of the power meter;
- the linearity (or log fidelity) of the signal generator due to the fact that the level setting of the generator in actual
measurement may be different from the level setting in the path compensation;
- change of the insertion loss between the path compensation and the actual measurement including repeatability
of the switches in the switch unit.
- error due to interpolation between correction factors at different frequencies.
The mismatch uncertainty is analysed in clause 6.10.4.1.2.
ETSI
175 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.10.4.1.2 Mismatch uncertainties

For the analysis of the overall mismatch uncertainty, firstly the external path compensation is analysed. The settings are
shown on figure 48. (The reason for introducing sub network D is that it is not necessarily the same sub network used in
the actual measurement):
Power
meter A
Sub network C
Generator Sub network D EUT conn. Power

Sub network B meter B
S4
Switch unit
Measurement 1
Power
meter A
Sub network C
Generator Sub network D EUT conn. Power

Sub network B meter B
S4
Switch unit
Measurement 2
Figure 48: The external path compensation
In the power meter B reading (measurement 1) the following mismatch uncertainties contribute to the reading:
Between Generator and sub network D
Between sub network D and switch S4
Between switch S4 and sub network B
Between sub network B and power meter B
Between Generator and switch S4 (through sub network D)
Between sub network D and sub network B
Between switch S4 and power meter B
Between Generator and sub network B (through sub network D)
Between sub network D and power meter B
Between Generator and power meter B (through sub network D)
ETSI
176 ETSI TR 100 028-1 V1.4.1 (2001-12)
In the power meter A reading (measurement 2) the following mismatch uncertainties contribute to the reading:
Between switch S4 and sub network C
Between sub network C and power meter A
Between Generator and switch S4 (through sub network D)
Between sub network D and sub network C
Between switch S4 and power meter A
Between Generator and sub network C (through sub network D)
Between sub network D and power meter A
Between Generator and power meter A (through sub network D)
As can be seen some of the mismatch uncertainties are part of both measurements (between Generator and switch S4),
so they cancel. The following mismatch uncertainties remain:
ETSI
177 ETSI TR 100 028-1 V1.4.1 (2001-12)
Some of these uncertainties will cancel later in the process. Then the internal path compensation (with settings as shown
in figure 49) is analysed:
Power
meter A
Sub network C
Generator Sub network A

EUT conn. EUT
Sub network B
S4
Switch unit
Measurement 3
Figure 49: The internal path compensation
In the internal path compensation (measurement 3) the following mismatch uncertainties contribute to the reading:
Between Generator and sub network A
Between sub network A and switch S4
Between Generator and switch S4 (through sub network A)
Between sub network A and sub network C
Between Generator and sub network C (through sub network A)
Between sub network A and power meter A
Between Generator and power meter A (through sub network A)
As can be seen, again, some of the mismatch uncertainties are part of both the internal and the external path
compensation (between switch S4 and power meter A), so they cancel.
The following mismatch uncertainties remain from the total path compensation:
ETSI
178 ETSI TR 100 028-1 V1.4.1 (2001-12)
Again, some of these uncertainties will cancel later in the process.
Then the actual measurement (with settings as shown in figure 50) is analysed:
Power
meter A
Sub network C
Generator Sub network A EUT conn. EUT

Sub network B
S4
Switch unit
Measurement 4
Figure 50: The actual measurement
In the actual measurement (measurement 4) the following mismatch uncertainties contribute:
Between sub network B and the EUT
Between sub network A and sub network B
Between switch S4 and the EUT
Between Generator and sub network B (through sub network A)
Between sub network A and the EUT
Between Generator and the EUT (through sub network A)
ETSI
179 ETSI TR 100 028-1 V1.4.1 (2001-12)
As can be seen, again, some of the mismatch uncertainties are part of both path compensation and the actual
measurement (between switch S4 and sub network B), so they cancel.
The following mismatch uncertainties remain in the total measurement:
Between sub network B and the EUT
Between sub network A and sub network B
Between switch S4 and the EUT
Between Generator and sub network B (through sub network A)
Between sub network A and the EUT
Between Generator and the EUT (through sub network A)
ETSI
180 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.10.4.2 Transmitter measurements

For the purpose of analysing receiver measurements the test system can be simplified as showed in figure 51:
Power
meter A
Sub network C
S8
Sub network F EUT
Sub network B EUT conn.
Signal analyser Sub network A
S5
Sub network D
Generator Sub network E
Sig. Gen. out
S9
Switch unit
Figure 51: Model for analysis of transmitter measurements
In figure 51:
Sub network A consists of all components and cables between the signal analyser and switch S5
Sub network B consists of all components and cables between switch S5 and switch S8
Sub network C consists of all components and cables between switch S5 and power meter A
Sub network D consists of all components and cables between switch S9 and switch S8
Sub network E consists of all components and cables between switch S9 and Generator
Sub network F consists of all components and cables between switch S8 and the EUT connector
Sub network G (not shown on this figure) consists of all components and cables between switch S9 and power meter B
(including an external cable and a 10 dB attenuator)
The external path compensation is performed as follows. This is not done in connection with every measurement, but
may be done with for instance 3 month intervals.
Measurement 1:
1) Power meter B is connected to the sig. gen. out connector through a cable and a 10 dB attenuator. Switch S9 is
set so the generator is connected to the sig. gen. out connector.
ETSI
181 ETSI TR 100 028-1 V1.4.1 (2001-12)
Measurement 2:
1) Then power meter B is removed and the 10 dB attenuator is connected to the EUT connector. Switch S5 and S8
are set so the generator is connected to power meter A through the 10 dB attenuator.
2) For all the frequencies and generator level settings in step 2 the power level is measured by power meter A. The
readings are stored by the test system
3) For each frequency point the correction factor is calculated as the difference (in dB) between the reading from
power meter A and power meter B. These are the external correction factors (path compensation data) stored by
the test system.
The internal path compensation is performed as follows. This is done prior to every measurement as an integral part of
the test case.
Measurement 3:
1) Switch S5, S8, and S9 are set so the generator is connected to power meter A through sub network D.
Measurement 4:
1) Then switch S5 is set so the generator is connected to the signal analyser.
2) For all the frequencies and generator level settings in step 2 the power level is measured by the signal analyser.
The readings are stored by the test system
3) For each frequency point the correction factor is calculated as the difference (in dB) between the reading from
the signal analyser and power meter A. These are the internal correction factors (path compensation data) stored
by the test system.
The actual test is performed as follows.
Measurement 5:
2) Switch S5 and S8 are set so the EUT is connected to the signal analyser
3) The relevant power level generated by the EUT is measured, and the reading from the signal analyser is stored
by the test system.
4) The final result is then calculated as the reading from step 3 (in dBm) minus the external and the internal
correction factors at the appropriate frequency from step 3. (If a correction factor at the measuring frequency
does not exist it is found by interpolation between the two correction factors at the nearest frequencies on each
side.)

The combined path compensation procedure and the actual test described consist of 5 individual measurements as
shown in figure 52 to figure 54: two measurements in external the path compensation part, two in the internal path
compensation and one in the actual measurement.
ETSI
182 ETSI TR 100 028-1 V1.4.1 (2001-12)
- The generator has an static error of Egen1 dB in measurement 1(compared to the setting of the generator level).
- Between measurement 1 and measurement 2 there is a generator drift dEgen1 dB.
- The generator has an static error of Egen2 dB in measurement 3(compared to the setting of the generator level).
- Between measurement 3 and measurement 4 there is a generator drift dEgen2 dB.
- The attenuation between the generator and the switch S9 is AttE1 dB in measurement 1.
- Between measurement 1 and measurement 2 there is an attenuation change in AttE1 of dAttE1 dB.
- The attenuation between the generator and the switch S9 is AttE2 dB in measurement 3.
- Between measurement 3 and measurement 4 there is an attenuation change in AttE2 of dAttE2 dB.
- The attenuation of switch S9 and sub network G is AttG dB in measurement 1.
- Between measurement 1 and measurement 2 there is an attenuation change in AttG of dAttG dB.
- The attenuation of switch S8 and sub network F is AttF dB in measurement 2.
- Between measurement 2 and measurement 5 there is an attenuation change in AttF of dAttF dB.
- The attenuation of switch S8, switch S9 and sub network D is AttD dB in measurement 3.
- Between measurement 3 and measurement 4 there is an attenuation change in AttD of dAttD dB.
- The attenuation of sub network B is AttB dB in measurement 2.
- Between measurement 2 and measurement 3 there is an attenuation change in AttB of dAttB1 dB.
- The attenuation between sub network B and power meter A is AttC dB in measurement 2.
- Between measurement 2 and measurement 3 there is an attenuation change in AttC of dAttC dB.
- The attenuation between sub network B and the signal analyser is AttA in measurement 4.
- Between measurement 4 and measurement 5 there is an attenuation change in AttA of dAttA dB.
- There is a static error of EpowA dB in power meter A in measurement 2.
- Between measurement 2 and measurement 3 there is a change in EpowA of dEpowA.
- There is a static error of EpowB dB in power meter B in measurement 1.
- The signal analyser error is Esa in measurement 4.
- Between measurement 4 and measurement 5 there is a drift and log fidelity error in Esa of dEsa.
- The EUT is generating a power level of Peut in the actual measurement.
In measurement 1 the reading from power meter B is:
P1 = Pgen1 + Egen1 – AttE1 – AttG + EpowB
P2 = Pgen1 + Egen1 + dEgen1 – AttE1 – dAttE1 – AttG –dAttG –AttF – AttB - AttC + EpowA
ETSI
183 ETSI TR 100 028-1 V1.4.1 (2001-12)
The external correction factor is:
Ccorr1 = P2 – P1 =
(Pgen1 + Egen1 + dEgen1 – AttE1 – dAttE1 – AttG –dAttG –AttF – AttB - AttC + EpowA) – (Pgen1 + Egen1 –
AttE1 – AttG + EpowB) =
dEgen1 – dAttE1 – dAttG –AttF – AttB - AttC + EpowA - EpowB
P3 = Pgen2 + Egen2-AttE2-AttD-dAttB- dAttB1 - AttC – dAttC + EpowA + dEpowA
In measurement 4 the reading from the signal analyser is:

P4 = Pgen2 + Egen2-AttE2-AttD-dAttB- dAttB2-AttA + Esa
The internal correction factor:
Ccorr2 = P4 – P3 =
Pgen2 + Egen2-AttE2-AttD-dAttB- dAttB2-AttA + Esa – (Pgen2 + Egen2-AttE2-AttD-dAttB- dAttB1 - AttC –

dAttC + EpowA + dEpowA) =
-dAttB- dAttB2-AttA + Esa + dAttB- dAttB1 + AttC + dAttC – EpowA-dEpowA
In measurement 5 (the actual measurement) the reading from the signal analyser is:
P5 = Peut – AttF – dAttF – AttB – dAttB3 – AttA – dAttA + Esa + dEsa
The result of the measurement is:
P5 – Ccorr1 – Ccorr2 =
(Peut – AttF – dAttF – AttB – dAttB3 – AttA – dAttA + Esa + dEsa) – (dEgen1 – dAttE1 – dAttG –AttF – AttB -
AttC + EpowA – EpowB) – (-dAttB- dAttB2-AttA + Esa +dAttB- dAttB1 + AttC + dAttC – EpowA-dEpowA) =
Peut – dAttF – dAttB3 – dAttA + dEsa – dEgen1 + dAttE1 + dAttG + EpowB + dAttB + dAttB2 –dAttB +
dAttB1 - dAttC + dEpowA
Again, as can be seen from the calculated result, again static errors in the combined measurement except the power
meter B error have cancelled. Apart from that only the drift and linearity errors remain.
- the absolute uncertainty of external power meter B
- the drift and linearity (or log fidelity) errors of the signal generator, the internal power meter A and the signal
analyser
- change of the insertion loss between the various measurements including repeatability of the switches in the
switch unit
The mismatch uncertainty is analysed in clause 6.10.4.2.2.
ETSI
184 ETSI TR 100 028-1 V1.4.1 (2001-12)
6.10.4.2.2 Mismatch uncertainties

For the analysis of the mismatch uncertainty, firstly the external path compensation is analysed.
It consists of two measurements, and the settings are shown in figure 52:
Power
Sub network C meter A
S8 Sub network F EUT conn.

Sub network B
Signal analyser Sub network A Power
meter B
S5
Sub network D
Sub network G
Sig. Gen. out

S9
Switch unit
Measurement 1
Power
S8 Sub network F EUT conn

Sub network B
S5
Sub network D
Sub network G
Sig. Gen. out

S9
Switch unit
Measurement 2
Figure 52: External path compensation
In the power meter B reading (measurement 1) the following mismatch uncertainties contribute:
Between Generator and sub network E
Between sub network E and switch S9
Between switch S9 and sub network G
ETSI
185 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between sub network G and power meter B
Between Generator and switch S9
Between sub network E and sub network G
Between Generator and sub network G (through sub network E)
Between sub network E and power meter B
Between Generator and power meter B
In the power meter A reading (measurement 2) the following mismatch uncertainties contribute:
Between switch S9 and sub network G
Between sub network G and sub network F
Between sub network F and switch S8
Between sub network B and switch S5
Between sub network E and sub network G
Between switch S9 and sub network F
Between sub network G and switch S8
Between sub network F and sub network B
Between switch S8 and switch S5
Between sub network B and sub network C
Between Generator and sub network G
Between sub network E and sub network F
Between sub network G and sub network B
Between sub network B and power meter A
Between Generator and sub network F
ETSI
186 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between sub network F and sub network C
Between sub network E and sub network B
Between sub network G and sub network C
Between sub network F and power meter A
Between Generator and sub network B
Between sub network G and power meter A
Between sub network E and sub network C
Between Generator and sub network C
Between sub network E and power meter A
Between Generator and power meter A
As can be seen some of the mismatch uncertainties are part of both measurements (between Generator and sub network
G), so they cancel.
The following mismatch uncertainties remain:
ETSI
187 ETSI TR 100 028-1 V1.4.1 (2001-12)
Then the analysis of the mismatch uncertainty from the internal path compensation is performed.
ETSI
188 ETSI TR 100 028-1 V1.4.1 (2001-12)
It also consists of two measurements, and the settings are shown in figure 53:
Power

Sub network B
S5
Sub network D
Sig. Gen. out

S9
Switch unit
Measurement 3
Power

Sub network B
S5
Sub network D
Sig. Gen. out

S9
Switch unit
Measurement 4
Figure 53: Internal path compensation
In the power meter A reading (measurement 3) the following mismatch uncertainties contribute:
Between switch S9 and sub network D
Between sub network E and sub network D
ETSI
189 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between Generator and sub network B)
Between Generator and power meter
In the signal analyser reading (measurement 4) the following mismatch uncertainties contribute:
Between switch S9 and sub network D
Between switch S5 and sub network A
Between sub network A and signal analyser
ETSI
190 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between sub network E and sub network D
Between sub network B and sub network A
Between switch S5 and signal analyser
Between sub network B and signal analyser
Between sub network D and sub network A
Between sub network D and signal analyser
Between sub network E and sub network A
Between sub network E and signal analyser
Between Generator and signal analyser
As can be seen, again some of the mismatch uncertainties are part of both measurements (all of them except where the
signal analyser, power meter A, sub network A, and sub network C is part), so they cancel.
ETSI
191 ETSI TR 100 028-1 V1.4.1 (2001-12)
Some of the remaining mismatch uncertainties contribute to both the external and the internal path compensation
(uncertainty components between switch S8 and power meter A)– therefore they also cancel. (When the two lists of
mismatch uncertainties are combined it is necessary to mark some of them with extra information in order to distinguish
between uncertainties which are between the same components, but with a different path between the two components.
For instance between the generator and power meter A)
The remaining uncertainties are:
Between switch S9 and sub network C (Through sub network D)
Between sub network E and sub network C (Through sub network D)
Between switch S9 and power meter A (Through sub network D)
ETSI
192 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between Generator and sub network C (Through sub network D)
Between sub network E and power meter A (Through sub network D)
Between Generator and power meter A (Through sub network D)
ETSI
193 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between switch S9 and sub network C (Through sub network G)
Between sub network E and sub network C (Through sub network G)
Between switch S9 and power meter A (Through sub network G)
Between Generator and sub network C (Through sub network G)
Between sub network E and power meter A (Through sub network G)
Between Generator and power meter A (Through sub network G)
Finally the analysis of the mismatch uncertainty from the actual measurement is performed.
The settings are shown in figure 54:
Power
S8 Sub network F EUT

Sub network B
Sub network A EUT conn
Signal analyser
S5
Sub network D
Sig. Gen. out

Switch unit S9
Measurement 5
Figure 54: The actual measurement
ETSI
194 ETSI TR 100 028-1 V1.4.1 (2001-12)
In the actual measurement (measurement 5) the following mismatch uncertainties contribute:
Between EUT and sub network F
Between EUT and switch S8
Between EUT and sub network B
Between sub network F and sub network A
Between EUT and sub network A
Between sub network F and signal analyser
Between EUT and signal analyser
As can be seen, again some of the mismatch uncertainties are part of both the actual measurements and the path
compensation (some components between switch S8 and the signal analyser), so they cancel.
Between switch S9 and sub network C (Through sub network D)
Between sub network E and sub network C (Through sub network D)
Between switch S9 and power meter A (Through sub network D)
Between Generator and sub network C (Through sub network D)
Between sub network E and power meter A (Through sub network D)
Between Generator and power meter A (Through sub network D)
ETSI
195 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between switch S9 and sub network C (Through sub network G)
Between sub network E and sub network C (Through sub network G)
Between switch S9 and power meter A (Through sub network G)
Between Generator and sub network C (Through sub network G)
Between sub network E and power meter A (Through sub network G)
ETSI
196 ETSI TR 100 028-1 V1.4.1 (2001-12)
Between Generator and power meter A (Through sub network G)
Between EUT and sub network F
Between EUT and sub network B
Between sub network F and sub network A
Between EUT and sub network A
Between sub network F and signal analyser
Between EUT and signal analyser
If there are for example 30 components involved in each measurement there are 5 times 435 = 2 175 mismatch
uncertainties involved before reduction. In some test systems there are even more components. This is the reason why
there can be several thousand mismatch uncertainties in a single measurement.
6.10.5 Summary
As mentioned earlier the individual components can be calculated when their individual losses and reflection
coefficients are known. The main problem is that some of the components are internal, so the relevant parameters can
not be measured directly without taking the switch unit apart.
The appropriate reflection coefficients may instead be assumed or calculated based on knowledge about the individual
components of the sub network.
In addition to the mismatch uncertainties derived previously, there may be others. For example in some receiver tests it
is necessary to switch attenuators in during the level settings because this gives a lower uncertainty than relying on the
linearity of the generators. This, however, adds to both the mismatch uncertainty and may add new power meter
linearity errors which must be taken into account.
As indicated the mismatch uncertainty calculation can be very complicated. Nevertheless it is necessary to perform the
calculations of the overall measurement uncertainty for a test performed on such a test system.
One way to simplify it is to use software tools which can actually handle all the (sometimes more than 100) components
in a test system. (Such a tool has actually been developed, but none are yet commercially available).
Such a tool must be capable of analysing networks with many components based on components data (s parameters),
the component's location in the network, and which other components it is connected to.
To calculate the over all mismatch uncertainty (as done above) it must calculate the uncertainties from the different
individual measurements and identify which uncertainties cancel.
Another simplified method could be to assume that cables and switches are loss-less when looking at mismatch
uncertainties, This results in a lot of errors being identical. It gives a little more conservative figure for the uncertainty
because the reduction in the mismatch due to loss between the two parts are not considered.
All of the listed uncertainties are between two sub networks, instruments or components which in some cases are
separated by other sub networks. If so the mismatch uncertainties are reduced due to insertion loss between the two
parts.
Further reductions can be accomplished by ignoring mismatch errors which are insignificant compared to the overall
mismatch uncertainty. If for instance if the two parts are separated by more than 10 dB they will be reduced by at least a
factor of 10. But care must be taken: some uncertainties may be caused by filters outside their pass bands causing their
reflection coefficients to be close to 1. These should not automatically be ignored as they would be significant even
with losses much greater than 10 dB involved.
The example in clause 6.10.5.1 shows how one of the mismatch uncertainties can be calculated if all the individual
components are known.
ETSI
197 ETSI TR 100 028-1 V1.4.1 (2001-12)
A third approach could be to estimate the s-parameters for the different logical parts of the test system as they are
shown on figure 51. This could be done by network analyser measurement on the switch unit connectors, by
measurements on internal connectors or by assuming internal s-parameters based on external measurements or
component data.
All the individual mismatch components (for instance the 55 components derived in clause 6.10.4.2.2) could be
programmed in a spreadsheet program, so it would be easy to input new sets of s-parameters representing other
frequencies or other switch unit settings.
6.10.5.1 Typical mismatch example

This example shows the calculation of the mismatch uncertainty between sub network C and sub network E through sub
network G (from measurement 2, figure 52) in the external path compensation procedure related to a transmitter
measurement.
Some details for the calculations must be assumed: Generator 2 is the generator, and the signal is connected through the
6 dB combiner and switch S10.
Furthermore it is assumed that attenuator 2 is by-passed between switches S6 and S7 during the path compensation and
the actual measurement. (See figure 46)
Then from figure 46 and figure 52 it can be seen that sub network C consists of a cable, switch S3, and a cable, and sub
network E consists of the cable connecting the generator to the switch unit, a cable, switch S11, a cable, the combiner, a
cable, switch S10, and a cable.
The loss separating these two sub networks consists of 3 switches and some cables, and a 10 dB attenuator.
In order to simplify the calculations it is assumed that the cables and the switches are loss-less. This will make the
calculations slightly more conservative since there will be no reduction of the mismatch uncertainty due to the loss
between the components, as loss between consists of only cables and switches.
The next assumption is that all cables are identical and all switches are identical.
Since a power combiner with 3 ports is involved, there will be a main path to be analysed, but in addition there will be
components from the 3. port of the power combiner as well with the same set of components as between generator 2
and the combiner.
Figure 55 applies for the calculations:
Generator Switch
Cable
cable S11
Power Switch Switch

Cable Cable Cable Cable
combiner S10 S3
Generator Switch SubnetworkE SubnetworkC

Cable
cable S12
Figure 55: The two sub networks in the mismatch uncertainty calculation
Each mismatch uncertainty component has one part on each side of the dashed line.
Since there is a power combiner with a loss of 6 dB involved, there will be components separated by 10 dB and
components separated by 16 dB as cables and switches are considered loss-less.
Since there are 10 components on the left side and 3 components on the right side there will be 30 mismatch
contribution (of which some will be identical).
ETSI
198 ETSI TR 100 028-1 V1.4.1 (2001-12)
They are:
1) 4 mismatch components due to mismatch between a cable and a cable, separated by 10 dB;
2) 1 mismatch component due to mismatch between a switch and a switch, separated by 10 dB;
3) 4 mismatch components due to mismatch between a switch and a cable, separated by 10 dB;
4) 2 mismatch components due to mismatch between a cable and the combiner, separated by 10 dB;
5) 1 mismatch component due to mismatch between a switch and the combiner, separated by 10 dB;
6) 4 mismatch components due to mismatch between a cable and a cable, separated by 16 dB;
7) 2 mismatch component due to mismatch between a switch and a switch, separated by 16 dB;
8) 6 mismatch components due to mismatch between a switch and a cable, separated by 16 dB;
9) 2 mismatch components due to mismatch between a switch and a generator cable, separated by 16 dB;
10) 4 mismatch components due to mismatch between a cable and a generator cable, separated by 16 dB.
The following data are assumed for the mismatch uncertainty calculations:
The reflection coefficient from a cable is 0,1
The reflection coefficient from a switch is 0,15
The reflection coefficient from the combiner is 0,08
The reflection coefficient from the 10 dB attenuator is 0,05
The reflection coefficient from a generator cable is 0,17
10 dB equals 0,3163 and 16 dB equals 0,1581
This gives the following standard deviation figures for the mismatch uncertainties:
1) 100 x 0,1 x 0,1/(0,3163 x 0,3163 x √2) %= 0,070 % 4 times;
2) 100 x 0,15 x 0,15/(0,3163 x 0,3163 x √2) %= 0,159 % 1 time;
3) 100 x 0,15 x 0,1/(0,3163 x 0,3163 x √2) %= 0,106 % 4 times;
4) 100 x 0,1 x 0,08/(0,3163 x 0,3163 x √2) %= 0,057 % 2 times;
5) 100 x 0,15 x 0,08/(0,3163 x 0,3163 x √2) %= 0,085 % 1 time;
6) 100 x 0,1 x 0,1/(0,1581 x 0,1581 x √2) %= 0,018 % 4 times;
7) 100 x 0,15 x 0,15/(0,1581 x 0,1581 x √2) %= 0,040 % 2 times;
8) 100 x 0,15 x 0,1/(0,1581 x 0,1581 x √2) %= 0,027 % 6 times;
9) 100 x 0,15 x 0,17/(0,1581 x 0,1581 x √2) %= 0,045 % 2 times;
10) 100 x 0,1 x 0,17/(0,1581 x 0,1581 x √2) %= 0,030 % 4 times.
This gives a total standard deviation = 0,34 % (≈ 0,03dB) calculated by applying the RSS method to the 30 uncertainty
components.
If only the components separated by more than 10 dB are considered, the result would be 0,32 % which is a little
smaller, but since the approach was conservative from the beginning it would be justified to do so.
A suitable way to do the calculations is to use a spread sheet program. calculations at different frequencies or with
changed components data can easily be done if the components data are entered so each component only need to be
modelled one time, which makes it much easier to re-do the analysis at different frequencies by just changing the
models data in the spread sheet.
ETSI
199 ETSI TR 100 028-1 V1.4.1 (2001-12)
All the individual uncertainty components are as usual combined as standard deviations as the square root of the sum of
the squares.
In a similar way the rest of the mismatch uncertainties can be analysed.
7 Transmitter measurement examples

The following pages of clause 7 show example measurement uncertainty calculations for a range of test configurations
involving a variety of uncertainty contributions. Components essential for the measurement uncertainty calculations are
shown in the accompanying drawings. Influence quantities (such as supply voltage and ambient temperature) are not
shown in the drawings although they are present in the examples.
Symbols and abbreviations used in the examples are explained in clauses 3.2 and 3.3. The test configuration,
uncertainty contributions and the calculations are only examples and may not include all the possibilities. It is important
that, where applicable, the errors are identified as either systematic or random for the purpose of making the
calculations. Each example is calculated for a confidence level of 95 %.
Many of the calculations on the following pages have been reproduced in spreadsheet form to provide the reader with a
structured and time-saving approach to calculating measurement uncertainty. The spreadsheets also allow the reader to
make modifications to the calculations to meet individual needs where the effects of each contribution can be assessed
more effectively. Where the related spreadsheet has been made available by ETSI, an appropriate reference has been
included in the text.
7.1 Conducted
7.1.1 Frequency error
a) Methodology
The signal to be measured is applied to a frequency counter via a power attenuator and the frequency read directly from
the counter (see figure 56).
Transmitter cable cable

Power Frequency
under counter
attenuator
test
Figure 56: Frequency error measurement configuration
For the purposes of this example the nominal frequency is assumed to be 900 MHz (uncertainty will be expressed as an
absolute value in Hz).
b) Measurement uncertainty
The time-base of the counter used has a drift of 1 × 10-9 per day. With a calibration period of less than 10 days, the time
base uncertainty is less then 1 × 10-8. The least significant digit is 10 Hz.
The manufacturers specification states that the overall uncertainty is time base uncertainty +3 counts of the least
significant digit or 30 Hz whichever is the greater. The uncertainty of the frequency counter related to the measurement
of 900 MHz is then:
- time base uncertainty = 900 × 106 × 1 × 10-8 = ±9 Hz (d) (r);
- counter uncertainty = 3 × 10 Hz = ±30 Hz (d) (r).
There is also an uncertainty associated with the ambient temperature uncertainty. The dependency values found in
table F.1 are:
- mean value of 0,02 ppm/°C;
- standard deviation of 0,01 ppm/°C.
ETSI
200 ETSI TR 100 028-1 V1.4.1 (2001-12)
This gives:
- 20 × 10-9 Hz/°C × 900 × 106 = 18 Hz/°C; and
- 10 × 10-9 Hz/°C × 900 × 106 = 9 Hz/°C.
Ambient temperature uncertainty is ± 3 °C (d) (r).
The standard uncertainty of the ambient temperature is:
3°C
u j tamb = = 1,73°C
3
The ambient temperature uncertainty is converted to a frequency uncertainty by means of formula 5.2.
u j ambient uncert ( 1,73 °C )2 × (( 18 Hz/ °C )2 +( 9 Hz/ °C )2 ) = 34,8 Hz
Finally the combined standard uncertainty is calculated:
(( 9 Hz )2 +( 30 Hz )2 )
u c frequency error= +( 34,8 Hz )2 = 39,2 Hz
3
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is
±1,96 × 39,2 Hz = ±76,8 Hz (see clause D.5.6.2 in TR 100 028-2 [8]).
c) Spreadsheet implementation of measurement uncertainty
This calculation has been implemented in a corresponding spreadsheet (see file "Frequency error.xls") and is available
in tr_10002801v010401p0.zip.
7.1.2 Carrier power

a) Methodology
The measurement is conducted using a power meter that consists of a thermocouple power sensor and a meter with a
built in reference source. A 30 dB power attenuator is used to reduce the level applied to the sensor (see figure 57).
Before making the measurement, the loss of the cable and power attenuator are determined using a signal generator.
The loss is measured by firstly connecting the generator to the power meter to obtain an arbitrary reference (see
figure 58). The cable and attenuator are then inserted between the sensor and generator and the loss determined (see
figure 59). Carrier power is recorded as the level measured on the power meter plus the measured correction for the
cable and attenuator.
Transmitter Cable 30dB power Power

attenuator Sensor
under test meter
Power ref
source
Figure 57: Carrier Power measurement configuration
NOTE 1: An additional example of carrier power uncertainty involving two attenuators can be found in clause 6.4.
i) Power meter and sensor:
power meter reference source level uncertainty is ±1,2 % (p) (d) (r).
1,2
u j reference level = = 0,03dB
3 × 23,0
ETSI
201 ETSI TR 100 028-1 V1.4.1 (2001-12)
Calibration factor uncertainty = ±2,3 % (p) (d) (r).
2,3
u j calibration factor = = 0,058dB
3 × 23,0
Range change error (one change) = ±0,5 % (p) (d) (r).
0,5
u j range change = = 0,013dB
3 × 23,0
Linearity factor = ±0,5 % (p) (d) (r).
0,5
u j linearity factor = = 0,013dB
3 × 23,0
Mismatch uncertainty when calibrating:
- reference source VSWRg = 1,05 (d) so reflection coefficient = 0,024;
- sensor input VSWRl = 1,15 (d) so reflection coefficient = 0,070.
0,024 × 0,07 × 100

u j calibration mismatch = = 0,01dB
2 ×11,5
Noise and drift is negligible at this power level and can be ignored.
u c power meter and sensor = u j ref level 2 + u j cal factor 2 + u j range change 2 + u j lin factor 2 + u j cal mismatch 2
u c power meter and sensor= 0,032 + 0,0582 + 0,0132 + 0,0132 + 0,012 = 0,069dB
ii) Uncertainty when measuring the attenuator/cable loss:
RF signal Power
generator Sensor
meter
Figure 58: Determining the reference level
RF signal Cable 30dB power Power

generator attenuator Sensor
meter
Figure 59: measuring the attenuator and cable loss
- generator reflection coefficient is 0,07 (d);
- sensor reflection coefficient is 0,07 (d);
- cable reflection coefficients are 0,14 (d);
- attenuator reflection coefficients are 0,13 (d).
For the calculation of mismatch uncertainty the cable attenuation is assumed to be 0 dB (x1 linear).
ETSI
202 ETSI TR 100 028-1 V1.4.1 (2001-12)
The mismatch uncertainty is comprised of two parts:
PART 1 - the reference measurement, this is when the generator and sensor are connected together (see figure 58):
0,07 × 0,07 × 100

u j mismatch: reference measurement = = 0,030dB
2 ×11,5
PART 2 - the attenuation measurement, this is when the cable and attenuator are placed between the generator and
sensor (see figure 59). Mismatch components appear at the generator/cable junction and the attenuator/sensor junction:
0,07 × 0,14 × 100

u j mismatch: generator to cable = = 0,060dB
2 ×11,5
0,13 × 0,07 × 100

u j mismatch: attenuator to sensor = = 0,056dB
2 ×11,5
0,07 × 0,13 × 12 × 100

u j mismatch: generator to attenuator = = 0,056dB
2 ×11,5
NOTE 2: The remaining mismatch contributions have a negligible effect due to the isolating effect of the 30dB
attenuator and have therefore been ignored.
The total mismatch uncertainty during the cable/attenuator measurement is:
u c mismatch: attenuation measurement = u j reference measurement

2
+u j generator to cable
2
+uj atten to sensor
2
+uj genertor to sensor
2
u c mismatch: attenuatio n measurement = 0,030 2 + 0,060 2 + 0,056 2 + 0,056 2 = 0,104 dB
The 30dB attenuator will reduce the level at the sensor so:
u c cable and attenuator measurement = u j range change error 2 + u j linearity factor 2 + u c mismatch: attenuation measurement
2
u c cable and attenuator measurement = 0,013 2 + 0,013 2 + 0,104 2 = 0,106 dB
iii) Carrier power measurement (see fig 44a):
Attenuator uncertainty:
Temperature influence = 0,0001 dB/degree which is negligible and can be ignored.
Power influence on the attenuator is 0,001 dB/dB x Watt (d) (r) = 0,001 x 30 x 25 = 0,75 dB (r):
0 ,75
u j power influence = = 0,433dB
3
- transmitter reflection coefficient is 0,5 (taken from table F.1);
- cable reflection coefficients are 0,14 (d);
- attenuator reflection coefficients are 0,13 (d);
- sensor reflection coefficient is 0,07 (d).
ETSI
203 ETSI TR 100 028-1 V1.4.1 (2001-12)
For the calculation of mismatch uncertainty the cable attenuation is assumed to be 0 dB (x 1 linear).
0,5 × 0,14 × 100%

u j mismatch: EUT to cable = = 0,430dB
2 × 11,5
0,5 × 0,13 × 12 × 100%

u j mismatch: EUT to attenuator = = 0,400dB
2 × 11,5
0,13 × 0,07 × 100%

u j mismatch: attenuator to sensor = = 0,056dB
2 × 11,5
The total mismatch uncertainty during the power measurement is:
u c mismatch: power measurement = u j EUT to cable 2 + u j EUT to attenuator 2 + u j attenuator to sensor 2
u c mismatch: power measuremen t = 0,430 2 + 0,400 2 + 0,056 2 = 0,60 dB
NOTE 3: The remaining mismatch contributions have a negligible effect due to the isolating effect of the 30dB
attenuator and have therefore been ignored.
Uncertainty due to influence quantities:
Ambient temperature = 20°C ± 1°C (d) (r).
The ambient temperature uncertainty is converted to a level uncertainty by means of formula 5.2 and table F.1.
Dependency values found in table F.1 are:
- mean of 4 % power/°C;
- standard deviation of 1,2 % power/°C.
Therefore:
 ( 1°C )2 
  × (( 4,0 % / °C )2 +( 1,2 % / °C )2 )
 3 

u j temperature uncertainty = = 0,1dB
23
Supply voltage = Vset ± 100 mV (d) (r).
The supply voltage uncertainty is converted to a level uncertainty by means of formula 5.2 and table F.1. Dependency
values found in table F.1 are:
- mean value of 10 % power/V;
- standard deviation of 3 % power/V.
Therefore:
 (0,1V )2 
  × ((10 % / V )2 + (3 % / V )2 )
 3 
 
u jsupply voltage uncertainty = = 0,026dB
23
Random uncertainty:
The measurement was repeated 9 times with the following results:
- 21,8 mW; 22,8 mW; 23,0 mW; 22,5 mW; 22,1 mW; 22,7 mW; 21,7 mW; 22,3 mW; 22,7 mW.
Mean value = 22,4 mW, standard deviation = 0,455 mW.
ETSI
204 ETSI TR 100 028-1 V1.4.1 (2001-12)
As the result is obtained as the mean value of 9 measurements the normalized standard deviation of the random
uncertainty is:
0,455mW
u i random = × 100 % = 0,03dB
22,4mW × 9 × 23,0
Uncertainty due to time duty cycle:
The standard uncertainty of the time duty cycle error (found in table F.1) = 2 % (p) (σ).
2%
u j time duty cycle = = 0,087dB
23,0
u c power measurement = u j pwr influence 2 + u c mismatch:Power measurement 2 + u j temp 2 + u j supply voltage

2
+ u i random 2 + u j time duty cycle
2
u c power measurement = 0,4332 + 0,6 2 + 0,12 + 0,026 2 + 0 ,032 + 0,087 2 = 0,75dB
The combined standard uncertainty for carrier power is:
u c carrier power = u c power meter & sensor 2 + u c cable & attenuator measurement 2 + u c power measurement 2
u c carrier power = 0,069 2 + 0,106 2 + 0,75 2 = 0,76dB
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 ×
0,76dB = ±1,49dB (see clause D.5.6.2 in TR 100 028-2 [8]).
This calculation has been implemented in a corresponding spreadsheet (see file "Carrier power_Rev1_V141.xls") and is
available in tr_10002801v010401p0.zip.
7.1.3 Adjacent channel power
7.1.3.1 Adjacent channel power method 1 (Using an adjacent channel power meter)
a) Methodology
The transmitter under test is connected to an adjacent channel power meter (power measuring receiver) via an attenuator
(see figure 60).
Modulating Low noise RF

AF oscillator signal generator
Transmitter cable Power

cable Power measuring
under
attenuator receiver
test
Figure 60: Measurement configuration for adjacent channel power measurement (method 1)
ETSI
205 ETSI TR 100 028-1 V1.4.1 (2001-12)
Power bandwidth of measuring receiver filter = ±0,2 dB (d) (r).
±0,20
u j filter bandwidth = = 0,115dB
3
Relative accuracy of measuring receiver = ±0,5 dB (d) (r).
±0,50
u j relative accuracy = = 0,289dB
3
Standard uncertainty of the random error = 0,11 dB (σ).
u i random = 0,11 dB
Deviation uncertainty = ±30 Hz (d) (r).
Deviation uncertainty is converted to a relative adjacent channel power uncertainty by means of formula 5.2 and
table F.1. Dependency values found in table F.1 are:
- mean value of 0,05 % (p) / Hz;
- standard deviation of 0,02 % (p) / Hz.
Therefore:



(30 Hz )2 
3 
(
× (0,05 % / Hz )2 + (0,02 % / Hz )2 )
 
u j converted deviation = =0,04dB
23,0
Uncertainty caused by measuring receiver filter position.
Uncertainty of 6 dB point = ±75 Hz (d) (r).
The uncertainty of the 6dB point is converted to a relative adjacent channel power uncertainty by means of formula 5.2
and table F.1. Dependency values found in table F1 are:
- mean value of 15 dB/kHz;
- standard deviation of 4 dB/kHz.
Therefore:
 ( 0 ,075 kHz )2 
u j converted filter position =   × (( 15 dB/kHz )2 +( 4 dB/kHz )2 ) = 0,67dB
 3 
 
u c a djacent channel power = u j filter pwr bw 2 + u j relative accuracy

2
+ ui random 2 + u j converted dev 2 + u j converted filter pos 2
u c adjacent channel power = 0,115 2 + 0,289 2 + 0,112 + 0,04 2 + 0,67 2 = 0,748dB
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × 0,748 dB
= ±1,47dB (see clause D.5.6.2 in TR 100 028-2 [8]).
This calculation has been implemented in a corresponding spreadsheet (see file "Adjacent channel power
(method 1)_V141.xls") and is available in tr_10002801v010401p0.zip.
ETSI
206 ETSI TR 100 028-1 V1.4.1 (2001-12)
7.1.3.2 Adjacent channel power method 2 (Using a spectrum analyser)

a) Methodology
The transmitter under test is connected to a spectrum analyser via a power attenuator (see figure 61) and the carrier is
recorded as reference.
Modulating
AF oscillator
Transmitter
cable 1 Power cable 2 Spectrum analyser
under
attenuator or selective voltmeter
test
Figure 61: Measurement configuration for adjacent channel power (method 2)
The adjacent channel power is calculated from the spectrum analyser readings (9 samples) by means of Simpson's Rule
(area under the curve).
Reference level (carrier power) uncertainty:
Spectrum analyser log fidelity = ±1 dB (d) (r) (carrier level may be measured below the analyser reference level).
±1,00
u j l og fidelity (ref level ) = = 0,577 dB
3
RBW switching = ±0,5 dB (d) (r).
±0,50
u jRBW switching = = 0,289 dB
3
Uncertainty of calculation caused by log fidelity (adjacent channel):
(The circles on figure 62 show the readings).
4 dB
3 dB
2 dB
1 dB
0 dB
-1 dB
-2 dB
-3 dB
- 4 dB
Figure 62: Typical screen view
ETSI
207 ETSI TR 100 028-1 V1.4.1 (2001-12)
Spectrum analyser log fidelity is a maximum of ±1,0 dB (d). Since the measured result is a sum of many contributions,
where the error can lie anywhere between ±1 dB, the combined error is assumed to be a Gaussian distribution, and the
±1,0 dB limits are assumed to be 3σ. The standard uncertainty is therefore 1/3 = 0,33 dB.
u j log fidelity (calculation ) = 0,33 dB
Random uncertainty:
Standard uncertainty of the random error is ± 0,11 dB (m) (σ).
ui random = 0,11dB
Deviation uncertainty:
Deviation uncertainty is ±30 Hz (d) (r).
Deviation uncertainty is converted to a level uncertainty by means of formula 5.2 and table F.1. Dependency values
found in table F1 are:
- mean of 0,05 % (p)/Hz;
- standard deviation of 0,02 % (p)/Hz.
Therefore:
 ( 30 Hz)2 
  × (( 0,05 %/Hz )2 +( 0,02 %/Hz )2 )
 3 
 
u jconverted deviation = = 0,04dB
23,0
Time-duty cycle:
Time-duty-cycle uncertainty (from table F.1): Standard deviation = 2,0 %(p).
2,0
u j TDC = = 0,087 dB
23,0
The combined standard uncertainty for adjacent channel power is:
uc adjacentchannelpower = u j logfidelity (ref level) 2 + u j RBW switching2 + u j logfidelity (calculation ) 2 + ui random2 + u j converteddeviation2 + u j TDC 2
u c adjacent channel power = 0,577 2 + 0,289 2 + 0,332 + 0,112 + 0,04 2 + 0,087 2 = 0,74dB
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × 0,74 dB
= ±1,45 dB (see clause D.5.6.2 in TR 100 028-2 [8]).
7.1.4 Conducted spurious emissions
7.1.4.1 Direct reading method

a) Methodology
A spectrum analyser is calibrated from its internal reference source using a cable with negligible loss at the calibration
reference frequency. The transmitter under test is then connected to the spectrum analyser via a 30 dB attenuator and
filter (see figure 63), and an absolute reading for each spurious emission obtained on the analyser. The levels are
corrected for attenuator loss, filter loss, and cable loss (which becomes significant at the higher spurious frequencies)
and recorded as the results for a direct reading. For this example, measurement uncertainty must include components of
uncertainty for the spectrum analyser, cable loss and various mismatches between the transmitter, cables, attenuator,
filter and spectrum analyser.
ETSI
208 ETSI TR 100 028-1 V1.4.1 (2001-12)
cable 30dB Spectrum analyser

Transmitter
power Filter or selective voltmeter
under test
attenuator
cal ref o/p
alternative cable
position when calibrating
Figure 63: Conducted spurious emission measurement configuration (direct method)
b) Measurement uncertainty: Direct method
Mismatch uncertainty when calibrating the spectrum analyser:
- spectrum analyser calibration reference output reflection coefficient is 0,2 (d);
- spectrum analyser RF input reflection coefficient is 0,1 (d);
- calibration cable reflection coefficient is 0,2 (m).
For calculation of mismatch, attenuation of the calibration cable is assumed to be 0,00dB (x1 linear).
0,2 × 0,2 × 100%

u j mismatch: calibration reference output and cable = = 2,828 % (v)
2
0,1 × 0,2 × 100%

u j mismatch: spectrum analyser input and cable = = 1,414 % (v)
2
0,1 × 0,2 × (1,0) 2 × 100%

u j mismatch: spectrum analyser input and spectrum analyser cal output = = 1,414 % (v)
2
The combined standard uncertainty for mismatch during calibration is:
u c mismatch: calibration = 2,828 2 + 1,414 2 + 1,414 2 = 3,464 % (v)
Mismatch uncertainty when measuring the transmitter spurious:
- transmitter reflection coefficient is 0,7 (from table F1);
- measurement cable reflection coefficients are 0,2 (m);
- filter reflection coefficients are 0,3 (d);
- spectrum analyser RF input reflection coefficient is 0,1 (d).
For the calculation of mismatch, measurement cable attenuation is assumed to be 0,00 dB (x1,0 linear) and filter
insertion loss is 1 dB (x 0,891 linear).
0,7 × 0,2 × 100 %

u jmismatch : transmitte r and cable = = 9,899 % ( v )
2
0, 2 × 0,1 × 100 %
u jmismatch : cable and attenuator = = 1,414 % (v )
2
0,1 × 0,3 × 100 %

u jmismatch : attenuator and filter = = 2,121 % (v )
2
0,3 × 0,1 × 100 %

u jmismatch : filter and spectrum analyser = = 2,121 % (v )
2
ETSI
209 ETSI TR 100 028-1 V1.4.1 (2001-12)
0,7 × 0,1 × (1,0) 2 × 100 %

u jmismatch : transmitte r and attenuator = = 4,950 % ( v )
2
0,1 × 0,1 × (0,891) 2 × 100 %

u jmismatch : attenuator and spectrum analyser = = 0,561 % (v )
2
uj mismatch: EUT and filter: Less than 0,01 % (v) due to the 30 dB attenuator, therefore neglected.
uj mismatch: EUT and spectrum analyser: Less than 0,01 % (v) due to the 30 dB attenuator, therefore neglected.
The combined standard uncertainty for mismatch with the transmitter connected is:
u c mismatch : transmitte r connected = 9,899 2 + 1, 414 2 + 2,121 2 + 2,121 2 + 4,950 2 + 0,561 2 = 11,567 % ( v )
The combined standard uncertainty for total mismatch is:
3,464 2 + 11,567 2
u c mismatch total : = = 1,05 dB
11,5
Uncertainty when making the measurement on the spectrum analyser:

Spectrum analyser calibration reference uncertainty = ±0,3 dB (d) (r).
0,3
u j cal ref = = 0,173dB
3
Spectrum analyser frequency response uncertainty = ±2,5 dB (d) (r).
2,5
u j frequency response = = 1,443dB
3
Spectrum analyser bandwidth switching uncertainty = ±0,5 dB (d) (r).
0,5
u j bandwidth switching = = 0,289dB
3
Spectrum analyser log fidelity = ±1,5 dB (d) (r).
1,5
u j lo g fidelity = = 0,866 dB
3
Spectrum analyser input attenuator switching uncertainty = ±0,2 dB (d) (r).
0,2
u j input att switching = = 0,115dB
3
Attenuator loss uncertainty = ±0,15 dB (d) (r).
0,15
u j atten loss = = 0,087dB
3
Filter loss uncertainty = ±0,15 dB (d) (r).
0,15
u j filter loss = = 0,087dB
3
Power coefficient of the attenuator = ±0,3 dB (c) (r).
0,3
u j att pwr coef = = 0,173dB
3
ETSI
210 ETSI TR 100 028-1 V1.4.1 (2001-12)
Standard uncertainty of measurement cable = ±0,2 dB (m) (σ).
u j cable = 0,2dB
Random uncertainty = ±0,2 dB (m) (σ).
u i random = 0,2dB
Uncertainty due to supply voltage:
Supply voltage uncertainty = ±100 mV (d) (r).
Supply voltage uncertainty is converted to a level uncertainty by means of formula 5.2 and table F.1. Dependency
- mean value of 10 % (p)/V;
- standard deviation of 3 % (p)/V.
Therefore:
 (0,1V )2 

 3
×

((10,0 %/V )2 + (3,0 %/V )2)
 
u j converted supply voltage = = 0,026dB
23,0
The combined standard uncertainty is:
u c tot = u jmiu2 + u jcal2 + u jfr 2 + u jbw 2 + u jlogf 2 + u jattsw 2 + u jattloss2 + u jfilter2 + u jatt p coef 2 + u jcable2 + u i rnd2 + u j vcc2
uc tot = 1.052 + 0,1732 + 1,4432 + 0,2892 + 0,8662 + 0,1152 + 0,0872 + 0,0872 + 0,1732 + 0,22 + 0,22 + 0,0262 = 2,05dB
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × 2,05 dB =
±4,02 dB (see clause D.5.6.2 in TR 100 028-2 [8]).
This calculation has been implemented in a corresponding spreadsheet (see file "Tx conducted spurious emissions
(direct)_V141.xls") and is available in tr_10002801v010401p0.zip.
7.1.4.2 Substitution method

a) Methodology
In order to reduce measurement uncertainty, a signal generator may be substituted for the transmitter and the level from
the generator increased until the same reading (as obtained with the transmitter) is measured on the analyser. The signal
generator output level is then recorded as the result using substitution. In this case, the large uncertainty of the spectrum
analyser is replaced with the much lower uncertainty of the signal generator, and the attenuator, filter and cable
uncertainties can be ignored since they are common to both measurements.
NOTE 1: In some cases the maximum signal generator level will be less than the transmitter spurious level, and the
substitution reading will be obtained from a different point on the spectrum analyser display (using the
analyser's dynamic range). For this reason the spectrum analyser log fidelity uncertainty has been
included in the calculation.
ETSI
211 ETSI TR 100 028-1 V1.4.1 (2001-12)
Signal
generator
alternative cable position
when substituting
30dB power
Filter Spectrum analyser
cable attenuator
Transmitter
under
test
Figure 64: Conducted spurious emissions measurement configuration (substitution method)
b) Measurement uncertainty: Substitution method
The 30 dB attenuator is large enough to provide good isolation between the transmitter (or signal generator) and the
filter. Thus the only mismatch uncertainty of interest is at the input to the attenuator. The rest cancel due to substitution:
- transmitter reflection coefficient is 0,7 (from table F1);
- measurement cable reflection coefficients are 0,2 (m);
- attenuator input reflection coefficient is 0,1 (d);
- signal generator output reflection coefficient is 0,35 (d).
For the calculation of mismatch, cable attenuation is assumed to be 0,00 dB (x 1 linear).
0,7 × 0,2 × 100%

u j mismatch: Tx to cable = = 9,899 % (v )
2
0,2 × 0,1 × 100%

u j mismatch: cable to attenuator = = 1,414 % (v)
2
0,7 × 0,1 × (1,0 2 ) × 100%

u j mismatch: Tx to attenuator = = 4,950 % (v)
2
0,35 × 0,2 × 100%

u j mismatch: sig gen to cable = = 4,950 % (v)
2
0,35 × 0,1 × (1,0 2 ) × 100%

u j mismatch: sig gen to attenuator = = 2,475 % (v)
2
The combined standard uncertainty for mismatch is:
9,899 2 + 1,414 2 + 4,950 2 + 4,950 2 + 2,4752

uc mismatch = = 1,083dB
11,5
Uncertainty when making the measurement:
Substitution signal generator level uncertainty is ±1 dB (d) (r).
±1,0
u j sig gen level = = 0,577dB
3
ETSI
212 ETSI TR 100 028-1 V1.4.1 (2001-12)
Spectrum analyser log fidelity (where signal generator is unable to produce sufficient level) = ±1,5 dB (d) (r).
1,5
u j lo g fidelity = = 0,866 dB
3
Random uncertainty is 0,2 dB (m) (σ).
Uncertainty due to supply voltage:
Supply voltage uncertainty = ±100 mV (d) (r).
Supply voltage uncertainty is converted to a level uncertainty by means of formula 5.2 and table F.1. Dependency
- mean value of 10 % (p)/V;
- standard deviation of 3 % (p)/V.
Therefore:


(0,1 V )2  × ((10 ,0% / V )2 +(3,0% / V )2 )
 3 
 
u jconverted supply voltage = = 0,026dB
23,0
u c conducted spurious emission = u c mismatch 2 + u j sig gen level 2 + u j log fidelity 2 + u i random 2 + u jsupply voltage uncert 2
u c conducted spuriousemissions = 1,0832 + 0,577 2 + 0,866 2 + 0,2 2 + 0,0262 = 1,52dB
±1,96 × 1,52 dB = ±2,98 dB (see clause D.5.6.2 in TR 100 028-2 [8]).
NOTE 2: The substitution example has a far lower measurement uncertainty than the direct example.
7.1.5 Intermodulation attenuation

directional
Transmitter 10dB power coupler <-1dB 20dB power Test signal
under test attenuator attenuator source
>-20dB
Power meter Spectrum 50 Ohm

sensor analyser termination
Figure 65: Intermodulation attenuation
The transmitter power is first measured on the power meter. The power meter is then connected to the 10 dB attenuator
(the connector which during the actual measurement is connected to the transmitter output) and the power meter reading
set to -30 dB (relative) by adjusting the level of test signal source. With the transmitter reconnected to the 10 dB
attenuator, the intermodulation component is then measured by direct observation on the spectrum analyser, and the
ratio of the largest intermodulation component to the carrier is recorded. As this is a relative measurement, uncertainties
due to the spectrum analyser (with the exception of log fidelity) cancel, and can be ignored.
ETSI
213 ETSI TR 100 028-1 V1.4.1 (2001-12)
b) Measurement uncertainty:
Uncertainty when measuring the transmitter level and setting the level of the test signal source to -30 dB relative:
NOTE: The power meter is only used to set the test signal source to -30 dB relative to the transmitter level so
only range change and linearity need to be considered.
Power meter linearity = ±0,5 % (p) (d) (r).
0,5
u j meter lin = = 0 ,013dB
3 × 23
Power meter range change error (one change) = ±0,5 % (p) (d) (r).
0,5
u j range change error = = 0 ,013dB
3 × 23
u c test signal level = 0,0132 + 0,0132 = 0 ,018dB
Mismatch uncertainty when measuring the transmitter level and setting the level of the test signal source to
-30 dB relative:
- transmitter reflection coefficient is 0,5 (table F.1);
- power sensor reflection coefficient is 0,07 (d);
- attenuator reflection coefficients are 0,1 (d) (both attenuators);
- directional coupler reflection coefficients are 0,05 (d).
For the following mismatch calculations the directional coupler loss is assumed to be 0 dB (x1 linear). The isolating
effect of the10 dB attenuator is however taken into consideration (multiplication by 0,316 in linear terms).
Only the reflection coefficients of the transmitter, 10 dB attenuator, the directional coupler and the 20 dB attenuator are
taken into account, the test signal source is ignored due to isolation. It is assumed that the spectrum analyser is
connected during the power measurement with the same cable and the same attenuator setting as during the
measurement. Therefore the mismatch uncertainties at this point cancel.
0,07 × 0,05 × 0,316 2 × 100%

u j mismatch: power sensor and directional coupler = = 0,025 % (v)
2
0,07 × 0,1× 0,316 2 × 12 × 100%

u j mismatch: power sensor and 20 dB attenuator = = 0 ,049 % (v)
2
u c mismatch: measuring test signal = 2,475 2 + 0,495 2 + 0,025 2 + 0,049 2 = 2,525 % (v)
Mismatch uncertainty with the transmitter reconnected to the 10dB attenuator:
Only the reflection coefficients of the transmitter, 10 dB attenuator, the directional coupler and the 20 dB attenuator are
taken into account, the test signal source is ignored due to isolation:
- transmitter reflection coefficient is 0,5 (table F1);
- directional coupler reflection coefficients are 0,05 (d).
ETSI
214 ETSI TR 100 028-1 V1.4.1 (2001-12)
For the following mismatch calculations the directional coupler loss is assumed to be 0 dB (x1 linear). The isolating
effect of the10 dB attenuator is however taken into consideration (multiplication with 0,316 in linear terms).
0 ,5 × 0,1 × 100%
u j mismatch: transmitter and 10dB attenuator = = 3,536 % (v )
2
0 ,5 × 0,05 × 0,316 2 × 100%

u j mismatch: transmitter and directional coupler = = 0,177 % (v)
2
0 ,5 × 0,1× 0,316 2 × 12 × 100%

u j mismatch : transmitter and 20 dB attenua tor = = 0,353% (v)
2
u c mismatch: test signal connected = 3,536 2 + 0 ,177 2 + 0,3532 = 3,558%(v)
Combined mismatch uncertainties:
2,5252 + 3,5582
uc mismatc h = = 0,379dB
11,5
Combined uncertainty of the test signal:
u c test signal = 0,0182 + 0,3792 = 0,379dB
Spectrum analyser log fidelity = ±1,5 dB (d) (r):
1,5
u j log fidelity = = 0,866dB
3
One of the intermodulation products has a 2nd order dependency from the unwanted signal corresponding to 2 dB/dB,
therefore the uncertainty of the level of the intermodulation product is doubled (see clause 6.5.5, and annex D
clauses D.3.4.5.2 and D.5).
The combined standard uncertainty for intermodulation attenuation is:
u c intermodulation attenuation measurement = 0 ,866 2 + (2 × 0,379)2 = 1,15dB
ETSI
215 ETSI TR 100 028-1 V1.4.1 (2001-12)
7.1.6 Attack time
7.1.6.1 Frequency behaviour (attack)

a) Methodology
Frequency behaviour (attack) is the time elapsed between switching on the transmitter and the moment when the carrier
frequency is within defined limits. Transmitter output frequency variation as a function of time during this period is
measured by means of a test discriminator providing vertical deflection to a storage oscilloscope (see figure 66).
Transmitter Trigger Storage

under device oscilloscope
test Tx on/off trigger
Y
input
cable Power cable Test

attenuator discriminator
Reference
signal
generator
Figure 66: Transmitter frequency/time measurement configuration (attack and release)
With the oscilloscope time base set to "repetitive" at an appropriate sweep rate, the oscilloscope display graticule is
calibrated by means of the signal generator, to provide vertical reference points corresponding to the specification
frequency limits or mask e.g. ± one channel. The oscilloscope is then set to "single sweep" in preparation for the
measurement.
When the trigger device is operated, it initiates the oscilloscope sweep and simultaneously switches on the transmitter.
Any variation in transmitter output frequency will appear at the discriminator output as a varying DC voltage which will
be recorded on the oscilloscope display as a plot of frequency against time.
- signal generator frequency uncertainty is ±10 Hz (d) (r);
- calibration uncertainty of discriminator (including the storage oscilloscope) is ±100 Hz (r);
- DC drift of discriminator is equivalent to ±100 Hz (d) (r).
Combined standard uncertainty of the frequency measurement:
(( 100 Hz )2 +( 100 Hz )2 +( 10 Hz )2 )
u j frequency measurement = = 81,9 Hz
3
Frequency uncertainty is converted to time uncertainty by means of formula 5.2 and table F.1. Dependency values
found in table F.1 are:
- mean value of 1,0 ms/kHz;
- standard deviation of 0,3 ms/kHz.
Therefore:
u j time = (0,0819 kHz ) 2 × ((1,0 ms/kHz ) 2 + (0,3 ms/kHz ) 2) = 0,086 ms
ETSI
216 ETSI TR 100 028-1 V1.4.1 (2001-12)
Random uncertainty is 0,5 ms (m) (c) (σ).
Oscilloscope timing uncertainty is ±1,0 ms (d) (r).
Trigger moment uncertainty is ±1,0 ms (d) (r).
The combined standard uncertainty:
 (1 ms )2 + (1 ms )2 
u c frequency behaviour = (0,086 ms )2 + (0,5 ms )2 +   = 0,961 ms
 3 
 
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × 0,961 ms =
±1,9 ms (see clause D.5.6.2 in TR 100 028-2 [8]).
7.1.6.2 Power behaviour (attack)

a) Methodology
Power behaviour (attack) is the time elapsed between switching on the transmitter and the moment when the transmitter
output power level is within defined limits i.e. a percentage of full power. Transmitter output power variation as a
function of time during this period is measured on a spectrum analyser set to zero span mode (see figure 67).
Transmitter Trigger trigger Spectrum

under device analyser
test Tx on/off
RF input
cable Power cable

attenuator
Figure 67: Transmitter power level/time measurement configuration (attack and release)
With the spectrum analyser time base set to "repetitive" at an appropriate sweep rate, the transmitter is switched on and
the analyser sensitivity adjusted until the measured signal coincides with the reference level. The analyser is then set to
"single shot", and the transmitter switched off in preparation for the measurement.
When the trigger device is operated, this simultaneously initiates the spectrum analyser sweep and switches on the
transmitter. Any variation in transmitter output power level will be recorded on the spectrum analyser display as a plot
of output power level against time.
Spectrum analyser log fidelity ±0,4 dB (d) (r).
0,4
uj log fidelity = = 0,231 dB
3
The power level difference uncertainty is then converted to time uncertainty by means of formula 5.2 and table F.1.
Dependency values found in table F.1 are:
- mean value of 0,3 ms/%;
- standard deviation of 0,1 ms/%.
Therefore:
u j time = (0,231× 11,5)2 × ((0,3 ms / %)2 + (0,1 ms / %)2 ) = 0,840 ms
Random uncertainty is 0,5 ms (m) (c) (σ).
ETSI
217 ETSI TR 100 028-1 V1.4.1 (2001-12)
Oscilloscope timing uncertainty is ±1,0 ms (d) (r).
Trigger moment uncertainty is ±1,0 ms (d) (r).
The combined standard uncertainty:
( 1 ms )2 +( 1 ms )2
u c frequency behaviour = ( 0,840 ms )2 +( 0,5 ms )2 +( ) = 1,274 ms
3
Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × 1,274 ms =
±2,5 ms (see clause D.5.6.2 in TR 100 028-2 [8]).
7.1.7 Release time
7.1.7.1 Frequency behaviour (release)

The only difference between this measurement and the measurement for attack in clause 7.1.6.1 is that in this case the
measurement is to determine the time elapsed between switching off the transmitter and the moment when the carrier
frequency falls outside defined limits. Measurement uncertainty for release is therefore the same as for attack.
7.1.7.2 Power behaviour (release)

The only difference between this measurement and the measurement for attack in clause 7.1.6.2 is that in this case the
measurement is to determine the time elapsed between switching off the transmitter and the moment when the carrier
power is within defined limits. Measurement uncertainty for release is therefore the same as for attack.
7.1.8 Transient behaviour of the transmitter

Transient behaviour of the transmitter is the period of transient frequency/power behaviour immediately following the
switching on or off of the transmitter.
7.1.8.1 Transient frequency behaviour

a) Methodology
Transient frequency behaviour is the frequency error of the transmitter during switch on and switch off transients.
Transmitter frequency error as a function of time during this period is measured by means of a test discriminator
providing vertical deflection to a storage oscilloscope (see figure 68).
Transmitter Trigger Storage

under device oscilloscope
test Tx on/off trigger
Y
input
cable Power cable Test

attenuator discriminator
Reference
signal
generator
Figure 68: Transmitter frequency/time measurement configuration (attack and release)
ETSI
218 ETSI TR 100 028-1 V1.4.1 (2001-12)
With the transmitter switched off, the oscilloscope time base is set to "repetitive" at an appropriate sweep rate. The
oscilloscope display graticule is then calibrated by means of the signal generator, to provide vertical reference points
corresponding to the specification frequency limits or mask e.g. ± one channel. When this has been accomplished, the
trigger selector is set to "single sweep" and the transmitter set to on or off depending upon which transient condition is
to be measured.
When the trigger device is operated, this simultaneously initiates the oscilloscope sweep and switches the transmitter on
or off according to the measurement. Any variation in transmitter output frequency will appear at the discriminator
output as a varying DC voltage which will be recorded on the oscilloscope display as a plot of frequency against time.
- signal generator frequency uncertainty is ±10 Hz (d) (r);
- calibration uncertainty of discriminator (including the storage oscilloscope) is ±100 Hz (d) (r);
- DC drift of discriminator is equivalent to ±100 Hz (d) (r).
The combined standard uncertainty of the frequency measurement:
(( 100 Hz )2 +( 100 Hz )2 +( 10 Hz )2 )
u c frequency measurement = = 81,9 Hz
3
±1,96 × 81,9 Hz = ±161 Hz (see clause D.5.6.2 in TR 100 028-2 [8]).
7.1.8.2 Power level slope

a) Methodology
Transmitter power output as a function of time (power level slope) is measured during switch on and switch off
transients by means of a spectrum analyser set to zero span mode (see figure 69).
Transmitter Trigger trigger Spectrum

under device analyser
test Tx on/off
RF input
cable Power cable

attenuator
Figure 69: Transmitter power level/time measurement configuration (attack and release)
With the transmitter switched on, and the spectrum analyser in zero span mode, the analyser sensitivity is adjusted until
the transmitter signal displayed on the screen coincides with the reference level. The trigger selector is then set to
"single shot", and the trigger device actuated to obtain a display of power level slope. The sweep is finally adjusted so
as to position the -6 dB point and the -30 dB points at left and right extremes of the display graticule, then the
transmitter switched on or off depending upon which transient condition is to be measured.
When the trigger device is operated, this simultaneously initiates the spectrum analyser sweep and switches the
transmitter on or off according to the measurement. Any variation in transmitter output power level will be recorded on
the spectrum analyser display as a plot of output power level against time.
ETSI
219 ETSI TR 100 028-1 V1.4.1 (2001-12)
(The following calculations are based on the assumption that the power level versus time is linear in logarithmic terms.)
Spectrum analyser log fidelity at -6 dB is ±0,6 dB (d) (r).
This is converted to time uncertainty: ±(0,6/(-6 + 30) × 100) % = ±2,5 %
Spectrum analyser log fidelity at -30 dB is ±1,5 dB (d) (r).
This is converted to time uncertainty: ±(1,5/(-6 + 30) × 100) % = ±6,25 %
Time measurement uncertainty (counts twice) is ±2 % of full screen ±2 % (d) (r).
Random uncertainty 1 % (m) (σ).
2,5 % 2 + 6 ,25 % 2+2,0 % 2+2,0 % 2

u c power level slope = +1,0 % 2 =4 ,33 %
3
±1,96 × 4,33 % = ±8,5 % (see clause D.5.6.2 in TR 100 028-2 [8]).
7.1.9 Frequency deviation
7.1.9.1 Maximum permissible frequency deviation

a) Methodology
The AF signal from the audio frequency oscillator is applied to the modulation input of the transmitter under test at a
level 20 dB above the level of normal test modulation (see figure 70).
Modulating
AF oscillator
Transmitter cable cable

Power Deviation
under
attenuator meter
test
Figure 70: Maximum permissible frequency deviation measurement configuration
The RF output from the transmitter under test is applied to a deviation meter through a power attenuator. The maximum
deviation is measured as 4,0 kHz.
As the modulating signal level is 20 dB above that required for normal test modulation, it is assumed that the AF level
uncertainty of the modulating AF oscillator has no influence.
Deviation uncertainty is ±1 % ±1 digit (f) (d) (r).
±1 digit is 10 Hz which is calculated as (10/4 000) × 100 % = ±0,25 %.
Residual modulation is ±20 Hz (f) (d) (r) which is converted to a percentage of the measured deviation (4 kHz):
(20/4 000) × 100 % = ±0,5 %
NOTE: The random contribution is deemed to be negligible and has therefore been ignored.
ETSI
220 ETSI TR 100 028-1 V1.4.1 (2001-12)
The combined standard uncertainty for maximum permissible frequency deviation is:
(1,0 % )2 + (0,25 % )2 + (0,5 % )2

u j total = = 0,66 %
3
±1,96 × 0,66 % = ±1,3 % (see clause D.5.6.2 in TR 100 028-2 [8]).
This calculation has been implemented in a corresponding spreadsheet (see file "Maximum permissible frequency
deviation_V141.xls") and is available in tr_10002801v010401p0.zip.
7.1.9.2 Response of the transmitter to modulation frequencies above 3 kHz

a) Methodology
The AF signal from the audio frequency oscillator is applied to the modulation input of the transmitter under test at the
specified level (see figure 71).
Modulating Low noise Audio

AF oscillator signal generator analyser
Transmitter
cable Power cable Deviation
under
attenuator meter
test
Figure 71: Measurement configuration for modulation frequencies above 3 kHz
The RF output from the transmitter under test is applied to a deviation meter through a power attenuator. The
demodulated signal is then applied to the audio analyser. A low noise signal generator is used as the local oscillator for
the deviation meter for demodulating signals with modulation frequencies above 3 kHz, to improve the noise behaviour.
The result is corrected for AF gain and AF filter shaping. It is assumed that the measurement is conducted sufficiently
above the measuring system noise level.
(As a low noise signal generator is used for the deviation meter local oscillator, it is assumed that residual deviation is
insignificant and has no influence on the measurement).
AF oscillator level uncertainty = ±0,70 % (v) (d) (r).
Deviation meter demodulator uncertainty = ±1,0 % (v) (d) (r).
Deviation meter AF gain uncertainty = ±2,0 % (v) (d) (r).
Audio analyser AC voltmeter uncertainty = ±4,0 % (v) (d) (r).
The combined standard uncertainty is then calculated:
( 0,70 % )2 +( 1,0 % )2+( 4,0 % )2 +( 2 ,0 % )2

uj = = 2,68%
3
The combined standard uncertainty is converted to dB: 2,68 %/11,5 = 0,233 dB.
ETSI
221 ETSI TR 100 028-1 V1.4.1 (2001-12)
This calculation has been implemented in a corresponding spreadsheet (see file "Response to mod freqs above
3kHz_V141.xls") and is available in tr_10002801v010401p0.zip.
7.2 Radiated tests

7.2.1 Frequency error (30 MHz to 1 000 MHz)
7.2.1.1 Anechoic Chamber

The method of calculating the expanded uncertainty for tests in which signal levels in dB are involved is equally
adopted for the frequency error test in which all the uncertainties are in the units of Hz. That is, all the uncertainty
contributions are converted into standard uncertainties and combined by the RSS method under the assumption that they
are all stochastic. All the uncertainty components which contribute to the test are listed in table 49. Annex A should be
consulted for the sources and/or magnitudes of the uncertainty contributions.
7.2.1.1.1 Contributions from the measurement
Table 49: Contributions from the measurement
uj or i Description of uncertainty contributions Hz

uj05 mutual coupling: detuning effect of the absorbing material on the EUT
uj09 mutual coupling: detuning effect of the test antenna on the EUT
sub-part 1, clause 5. The combined standard uncertainty of the frequency measurement
(uc contributions from the measurement) is the combination of the components outlined above.
uc = uc contributions from the measurement = __,__ Hz
7.2.1.1.2 Expanded uncertainty

±1,96 × uc = ±__,__ Hz (see clause D.5.6.2 in TR 100 028-2 [8]).
7.2.1.2 Anechoic Chamber with a ground plane


uj05 mutual coupling: detuning effect of the absorbing material on the EUT
ETSI
222 ETSI TR 100 028-1 V1.4.1 (2001-12)

sub-part 1, clause 5. The combined standard uncertainty of the frequency measurement (uc contributions from the
measurement) is the combination of the components outlined above.
7.2.1.3 Open Area Test Site



sub-part 1, clause 5. The combined standard uncertainty of the frequency measurement
(uc contributions from the measurement) is the combination of the components outlined above.
7.2.1.4 Stripline
This test is not usually performed in a Stripline and is therefore not considered here.
7.2.1.5 Test fixture

ETSI
223 ETSI TR 100 028-1 V1.4.1 (2001-12)

sub-part 1, clause 5. The combined standard uncertainty of the frequency measurement (uc contributions from the
measurement) is the combination of the components outlined above.

Using an expansion factor (coverage factor) of k = 1,96, the expanded measurement uncertainty is ±1,96 × uc =
±__,__ Hz (see clause D.5.6.2 in TR 100 028-2 [8]).
7.2.2 Effective radiated power (30 MHz to 1 000 MHz)

A fully worked example illustrating the methodology to be used can be found in TR 102 273 [3], part 1, sub-part 2,
clause 4.
7.2.2.1.1 Uncertainty contributions: Stage one: EUT measurement

For the measurement of effective radiated power two stages of test are involved. The first stage (the EUT measurement)
is to measure on the receiving device, a level from the EUT as shown in figure 72 (shaded components are common to
both stages of the test).
Test antenna
Test
antenna
cable 2
10 dB device
ferrite beads
Figure 72: Stage one: EUT measurement
mismatch uncertainty contributes identically in each stage and hence cancels. Similarly, the systematic uncertainty
contributions (e.g. test antenna cable loss, etc.) of the individual components also cancel.
The magnitude of the random uncertainty contribution to each stage of the procedure can be assessed from multiple
repetition of the EUT measurement. All the uncertainty components which contribute to this stage of the test are listed
in table 53. Annex A should be consulted for the sources and/or magnitudes of the uncertainty contributions.
ETSI
224 ETSI TR 100 028-1 V1.4.1 (2001-12)

uj19 cable factor: test antenna cable
uj47 receiving device: absolute level
uj06 mutual coupling: test antenna to its images in the absorbing material
measurement in dB.
7.2.2.1.2 Uncertainty contributions: Stage two: Substitution

The second stage (the substitution) involves replacing the EUT with a substitution antenna and signal source as shown
in figure 73 and adjusting the output level of the signal generator until the same level as in stage one is achieved on the
receiving device.
Test antenna
Test
antenna
Figure 73: Stage two: Substitution measurement
All the uncertainty components which contribute to this stage of the test are listed in table 54. Annex A should be
ETSI
225 ETSI TR 100 028-1 V1.4.1 (2001-12)

uj19 cable factor: substitution antenna cable
uj41 insertion loss: substitution antenna cable
uj40 insertion loss: substitution antenna attenuator
uj45 antenna: gain of the substitution antenna 0,50
uj22 position of the phase centre: substitution antenna
uj06 mutual coupling: substitution antenna to its images in the absorbing material
sub-part 1, clause 5. This gives the combined standard uncertainty (uc contributions from the substitution) for the substitution
measurement in dB.

The combined standard uncertainty of the effective radiated power measurement is the RSS combination of the
components outlined in clauses 7.2.2.1.1 and 7.2.2.1.2. The components to be combined are uc contribution from the EUT
measurement and uc contribution from the substitution.
uc = uc2contribution fromthe EUT measurement + uc2contribution fromthe substitution = __, __ dB
±__,__ dB (see clause D.5.6.2 in TR 100 028-2 [8]).
ETSI
226 ETSI TR 100 028-1 V1.4.1 (2001-12)

Test antenna
Test
antenna
cable 2
10 dB device
ferrite beads
Ground plane
mismatch uncertainty contributes identically to both stages and hence cancels. Similarly, the systematic uncertainty
The magnitude of the random uncertainty contribution to this stage of the procedure can be assessed from multiple
repetition of the EUT measurement. All the uncertainty components which contribute to this stage of the test are listed

uj16 range length
uj14 mutual coupling: test antenna to its image in the ground plane
measurement in dB.
ETSI
227 ETSI TR 100 028-1 V1.4.1 (2001-12)
7.2.2.2.2 Uncertainty contributions: Stage two: Substitution measurement

receiving device.
Test antenna
Test
antenna
Ground plane

uj02 reflectivity of absorbing material: substitution antenna to the test antenna
uj45 antenna: gain of substitution antenna
uj46 antenna: tuning of the substitution antenna
uj14 mutual coupling: substitution antenna to its image in the ground plane
uj11 mutual coupling: substitution antenna to the test antenna
measurement in dB.
ETSI
228 ETSI TR 100 028-1 V1.4.1 (2001-12)

components outlined in clauses 7.2.4.1 and 7.2.4.2. The components to be combined are uc contribution from the EUT
±1,96 × uc = ±__,__ dB (see clause D.5.6.2 in TR 100 028-2 [8]).

Test antenna
Test
antenna
cable 2
10 dB device
ferrite beads
Ground plane
repetition of the EUT measurements. All the uncertainty components which contribute to this stage of the test are listed
ETSI
229 ETSI TR 100 028-1 V1.4.1 (2001-12)

uj16 range length
measurement in dB.

receiving device.
Test antenna
Test
antenna
Ground plane
Figure 77: Stage two: Substitution
ETSI
230 ETSI TR 100 028-1 V1.4.1 (2001-12)

uj45 antenna: gain of the substitution antenna
measurement in dB.

7.2.2.4 Stripline

The uncertainty contributions for the test are shown in table 59.
ETSI
231 ETSI TR 100 028-1 V1.4.1 (2001-12)

The standard uncertainties from table 59 should be given values according to annex A. They should then be combined
by the RSS (root sum of the squares) method in accordance with TR 102 273 [3], part 1, sub-part 1, clause 5. This gives
the combined standard uncertainty (uc contributions from the measurement) for the EUT measurement in dB.

Tests in a Test Fixture differ to radiated tests on all other types of site in that there is only one stage to the test.
However, to calculate the measurement uncertainty, the Test Fixture measurement should be considered as stage two of
a test in which stage one was on an accredited Free-Field Test Site. The combined standard uncertainty, uc, of the
effective radiated power measurement is therefore, simply the RSS combination of the value
for uc contributions from the measurement derived above and the combined uncertainty of the Free-field Test Site
uc contribution from the Free-Field Test Site.
uc = uc2 contributions from the measurement + uc2 contributions from the free− field test site = __, __ dB
7.2.3 Radiated spurious emissions

For the measurement of spurious effective radiated power two stages of test are involved. The first stage (the EUT
measurement) is to measure on the receiving device, a level from the EUT as shown in figure 78 (shaded components
are common to both stages of the test).
Test antenna
Test
antenna
cable 2 Receiving
EUT Attenuator 2
10 dB device
ferrite beads
ETSI
232 ETSI TR 100 028-1 V1.4.1 (2001-12)
repetition of the EUT measurement.

uj54 EUT: influence of setting the power supply on the spurious emission level 0,03
uj51 EUT: influence of the ambient temperature on the spurious emission level 0,03
measurement in dB.
7.2.3.1.2 Uncertainty contributions: Stage two: Substitution

receiving device.
Test antenna
Test
antenna
ETSI
233 ETSI TR 100 028-1 V1.4.1 (2001-12)

sub-part 1, clause 5. This gives the combined standard uncertainty (uc contribution from the substitution) for the EUT
measurement in dB.

The combined standard uncertainty of the ERP measurement of the spurious emission is the combination of the
uc = uc2contribution fronthe EUT measurement + uc2contribtution fromthe substitution = __, __dB

ETSI
234 ETSI TR 100 028-1 V1.4.1 (2001-12)
Test antenna
Test
antenna
cable 2
10 dB device
ferrite beads
Ground plane
mismatch uncertainty contributes identically to both stages and hence cancels. Similarly, the systematic uncertainty

uj54 EUT: influence of setting the power supply on the spurious emission levels
uj16 range length
measurement in dB.
ETSI
235 ETSI TR 100 028-1 V1.4.1 (2001-12)

receiving device.
Test antenna
Test
antenna
cable 1 cable 2
Signal Attenuator 1 Attenuator 2 Receiving
Ground plane

uj02 reflectivity of absorbing material: substitution antenna to the test antenna
uj45 antenna: gain of substitution antenna
measurement in dB.
ETSI
236 ETSI TR 100 028-1 V1.4.1 (2001-12)

uc = uc2contribution fronthe EUT measurement + uc2contribtution fromthe substitution = __, __dB

Test antenna
Test
antenna
cable 2
10 dB device
ferrite beads
Ground plane
ETSI
237 ETSI TR 100 028-1 V1.4.1 (2001-12)

uj54 EUT: influence of setting the power supply on the spurious emission level
uj16 range length
uj13 mutual coupling: EUT to its images in the ground plane
uj14 mutual coupling: test antenna to its images in the ground plane
measurement in dB.

receiving device.
Test antenna
Test
antenna
cable 1 cable 2
Signal Attenuator 1 Attenuator 2 Receiving
Ground plane
Figure 83: Stage two: Typical emission substitution test
consulted for the sources and/or magnitudes of the uncertainty contribution.
ETSI
238 ETSI TR 100 028-1 V1.4.1 (2001-12)

measurement in dB.

components outlined in clauses 7.2.6.1 and 7.2.6.2. The components to be combined are uc contribution from the EUT
uc = uc2contribution fronthe EUT mesurement + uc2contribtution fromthe substitution = __, __dB
7.2.3.4 Stripline

This test is not normally carried out in a test fixture.
7.2.4 Adjacent channel power

This test is normally carried out using a test fixture and as a result has not been considered for the Anechoic Chamber.
ETSI
239 ETSI TR 100 028-1 V1.4.1 (2001-12)

This test is normally carried out using a test fixture and as a result has not been considered for the Anechoic Chamber
with a ground plane.

This test is normally carried out using a test fixture and as a result has not been considered for the Open Area Test Site.
7.2.4.4 Stripline
This test is normally carried out using a test fixture and as a result has not been considered for the Strip line.

The uncertainty contributions for the test are shown in table 66.
NOTE: Some standards require the adjacent channel power to be 60 dBc without the need for it to fall below
250 nW. In this case, both values (absolute and dBc) are required as, for example, 40 dBc is considered
satisfactory if the adjacent channel power is < 250 nW.

uj49 receiving device: power measuring receiver
The standard uncertainties from table 66 should be given values according to annex A. They should then be combined
by RSS in accordance with TR 102 273 [3], part 1, sub-part 1, clause 5. This gives the combined standard uncertainty
(uc contributions from the measurement) for the EUT measurement in dB.

For a relative measurement (dBc) of adjacent channel power, the combined uncertainty, uc, of the measurement is
simply the value for uc contributions from the measurement derived above.
For those test standards that require the adjacent channel power to be given in absolute terms, however, for the
calculation of the measurement uncertainty, the Test Fixture measurement should be considered as stage two of a test in
which stage one was on an accredited Free-Field Test Site. The combined standard uncertainty, uc, of the adjacent
channel power measurement is therefore, simply the RSS combination of the value for uc contributions from the measurement
derived above and the combined uncertainty of the Free-field Test Site uc contribution from the Free-Field Test Site.
uc = uc2 contributions from the measurement + uc2 contributions from the free− field test site = __, __ dB
ETSI
240 ETSI TR 100 028-1 V1.4.1 (2001-12)
Annex A:
Bibliography
• The new IEEE standard dictionary of electrical and electronic terms. Fifth edition, IEEE Piscataway, NJ USA
1993.
• Antenna theory, C. Balanis, J. E. Wiley 1982.
• Antenna engineering handbook, R. C. Johnson, H. Jasik.
• Control of errors on Open Area Test Sites , A. A. Smith Jnr. EMC technology October 1982 pg 50-58.
• IEC 60050-161: "International Electrotechnical Vocabulary. Chapter 161: Electromagnetic compatibility".
• The gain resistance product of the half-wave dipole, W. Scott Bennet Proceedings of IEEE vol. 72 No. 2
Dec 1984 pp 1824-1826.
• Wave transmission, F. R. Conner, Arnold 1978.
• Antennas, John D. Kraus, Second edition, McGraw Hill.
• Antennas and radio wave propagation, R. E. Collin, McGraw Hill.
ETSI
241 ETSI TR 100 028-1 V1.4.1 (2001-12)
History
Document history
Edition 1 March 1992 Publication as ETR 028
Edition 2 March 1994 Publication as ETR 028
V1.3.1 March 2001 Publication
V1.4.1 December 2001 Publication
ETSI

TR 10002801v010401p

Uploaded by

Copyright:

Available Formats

TR 10002801v010401p

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TR 10002801v010401p

Uploaded by

Copyright:

Available Formats

ETSI TR 100 028-1 V1.4.

650 Route des Lucioles

Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16

Siret N° 348 623 562 00017 - NAF 742 C

Individual copies of the present document can be downloaded from:

No part may be reproduced except as authorized by written permission.

© European Telecommunications Standards Institute 2001.

6.8.4.1 Uncertainty contributions: Stage 1: Transform Factor measurement..................................................109

6.10.4.2.2 Mismatch uncertainties..................................................................................................................184

7.2.3.1.2 Uncertainty contributions: Stage two: Substitution .......................................................................232

Annex A: Bibliography ........................................................................................................................240

Intellectual Property Rights

More precisely, the basic purpose of the present document is to:

It covers the following aspects relating to measurements:

b) recommended maximum acceptable uncertainties for each of the measured parameters;

c) a method of applying the uncertainties in the interpretation of the results.

[1] Guide to the Expression of Uncertainty in Measurement (International Organization for

[4] ITU-T Recommendation O.41: "Psophometer for use on telephone-type circuits".

[7] EN 55020: "Electromagnetic immunity of broadcast receivers and associated equipment".

3 Definitions, symbols and abbreviations

error (relative): ratio of an error to the true value

A practical form of this formula is:

- the same method of measurement;

- the same observer;

- the same measuring instrument;

- the same location;

- the same conditions of use;

- repetition over a short period of time

measurand: quantity subjected to measurement

psophometric weighting network: Should be as described in ITU-T Recommendation O.41

uncertainty (limits of uncertainty of a measuring instrument): extreme values of uncertainty permitted by

NOTE: This term is also known as "tolerance".

NOTE: It is the standard deviation of the corresponding distribution.

uj16 range length

4 Introduction to measurement uncertainty

4.1 Background to measurement uncertainty

Absolute error = the measured value - the true value.

This is further developed in clause D.5.6 of TR 100 028-2 [8].

4.1.2 Assessment of upper and lower uncertainty bounds

- 95 % confidence within ±1,96 × uc, etc.

4.1.3 Combination of rectangular distributions

Probability of a number (1 die) Probability of a number (2 dice)

Figure 1: One and two die outcomes

Probability of a number (3 dice) Probability of a number (4 dice)

Figure 2: Three and four die outcomes

Probability of a number (5 dice) Probability of a number (6 dice)

Figure 3: Five and six die outcomes

4.1.4 Main contributors to uncertainty

4.1.5 Other contributors

4.2 Evaluation of individual uncertainty components

- type A: those which are evaluated by statistical methods;

- type B: those which are evaluated by other means.

4.2.1 Evaluation of type A uncertainties

- What to do with the results?

- How much variation is acceptable?

- When do we suspect the measuring system is faulty?

- Are the conditions repeatable?

- Are the results stable with ambient conditions?

4.2.2 Evaluation of type B uncertainties