PDF PDF

An Interactive Approach to Signals and
Systems Laboratory
By:
Nasser Kehtarnavaz
Philipos Loizou
Mohammad Rahman
An Interactive Approach to Signals and
Systems Laboratory
By:
Nasser Kehtarnavaz
Philipos Loizou
Mohammad Rahman
Online:
< http://cnx.org/content/col10667/1.14/ >
CONNEXIONS
Rice University, Houston, Texas

This selection and arrangement of content as a collection is copyrighted by Nasser Kehtarnavaz, Philipos Loizou, Mo-
hammad Rahman. It is licensed under the Creative Commons Attribution 3.0 license (http://creativecommons.org/licenses/by/3.0/).
Collection structure revised: September 6, 2012
PDF generated: September 6, 2012
For copyright and attribution information for the modules contained in this collection, see p. 197.
Table of Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 LabVIEW Programming Environment
1.1 LabVIEW Programming Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Lab 1: Introduction to LabVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2 LabVIEW MathScript and Hybrid Programming

2.1 LabVIEW MathScript and Hybrid Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.2 Lab 2: LabVIEW MathScript and Hybrid Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3 Convolution and Linear Time-Invariant Systems

3.1 Convolution and Linear Time-Invariant Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2 Lab 3: Convolution and Its Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4 Fourier Series
4.1 Fourier Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.2 Lab 4: Fourier Series and Its Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5 Continuous-Time Fourier Transform

5.1 Continuous-Time Fourier Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.2 Lab 5: CTFT and Its Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6 Digital Signals and Their Transforms

6.1 Digital Signals and Their Transforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.2 Lab 6: Analog-to-Digital Conversion, DTFT and DFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
7 Analysis of Analog and Digital Systems

7.1 Analysis of Analog and Digital Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 175
7.2 Lab 7: System Response, Analog and Digital Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Attributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
iv
1
Preface
A typical undergraduate electrical engineering curriculum includes a signals and systems course during
which students are initially exposed to signal processing concepts such as convolution, Fourier series, Fourier
transform and ltering. Laboratory components of signals and systems courses are primarily based on textual
.m les. Although the ability to write textual codes is an important aspect of a lab component, students can
enhance their understanding of signal processing concepts in these courses if they interactively experiment
with their codes.
Our motivation for writing this book has thus been to present an interactive programming approach as
an alternative to the commonly practiced textual programming in signals and systems labs to provide an
ecient way for students to interact and experiment with their codes. The interactivity achieved via hybrid
programming, that is, a combination of textual and graphical programming, oers students a more eective
tool to better understand signal processing concepts.
Textual programming and graphical programming both have pros and cons. In general, math operations
are easier to code in textual mode. On the other hand, graphical programming oers an easy-to-build inter-
active and visualization environment along with a more intuitive approach toward building signal processing
systems.
To bring together the preferred features of textual and graphical programming, we have designed the labs
associated with a typical signals and systems course by incorporating .m les into the National Instruments
LabVIEW graphical programming environment. This way, although students program the code in textual .m
les, they can easily achieve interactivity and visualization in LabVIEW by just having some basic knowledge
of the software. The rst two labs provide an introduction to LabVIEW and MathScript (.m les) to help
students become familiar with both graphical and textual programming in case they have not already done
so in their earlier courses.
In addition to the signal processing concepts, students cover example applications in each lab to learn
how to relate concepts to actual real-world applications. The applications considered span dierent sig-
nal processing areas including speech processing, telecommunications and digital music synthesis. These
applications provide further incentive for students to stay engaged in the labs.
The chapters in this book are organized into the following labs:
1. Introduction to LabVIEW
Students gain some basic familiarity with LabVIEW, such as how to use controls, indicators and other
LabVIEW graphical features, to make .m les more interactive.
2. Introduction to MathScript
If not already familiar with .m le coding, students learn the basics of this coding.
3. Convolution and Linear Time-Invariant Systems
Students experiment with convolution and linear time-invariant (LTI) systems. Due to the discrete-time
nature of programming, students must make an approximation of the convolution integral. The lab, which
covers convolution properties, shows how to perform numerical approximation of convolution. To apply
convolution concepts, students examine an RLC circuit, and build and analyze an echo cancellation system.
4. Fourier Series and Its Applications
1 This content is available online at <http://cnx.org/content/m19826/1.1/>.
1
2
Students explore the representation of periodic analog signals using Fourier series and discuss the decom-
position and reconstruction of periodic signals using a nite number of Fourier coecients. To apply the
concepts they have learned, students perform an RLC circuit analysis using periodic input signals.
5. Continuous-Time Fourier Transform and Its Applications
Students implement continuous-time Fourier transform (CTFT) and its properties, as well as cover am-
plitude modulation and high-frequency noise removal as CTFT applications.
6. Digital Signals and Their Transforms
Students explore the transforms of digital signals. In the rst part of the lab, students examine analog-to-
digital conversion and related issues including sampling and aliasing. In the second part, students cover the
transformations consisting of discrete Fourier transform (DFT) and discrete-time Fourier transform (DTFT)
and compare them to the corresponding transforms for continuous-time signals, namely Fourier series and
CTFT, respectively. Students also examine applications such as dual-tone multi-frequency (DTMF) signaling
for touch-tone telephones and dithering to decrease signal distortion due to digitization.
7. Analysis of Analog and Digital Systems
During the nal lab, students implement the techniques and mathematical transforms they learned in
the previous labs to perform analog and digital ltering. They build and analyze a square root system and
a ltering system with interactive capabilities.
The codes and les associated with the labs in this book can be downloaded from the website at
www.utdallas.edu/∼kehtar/signals-systems(username = signals-systems, password = labora-
tory). Note that this book is meant only as an accompanying lab book to signals and systems textbooks
and should not be used as a substitute for these textbooks.
We would like to express our gratitude to National Instruments, in particular its Academic Marketing
Division and Mr. Erik Luther, for their support and initial publication of this book through lulu.com. We
hope its publication now through Connexions would facilitate its widespread use in signals and systems
laboratory courses.
Nasser Kehtarnavaz
Philipos C. Loizou
Mohammad T. Rahman
Chapter 1
LabVIEW Programming Environment
3
4 CHAPTER 1. LABVIEW PROGRAMMING ENVIRONMENT
1
1.1 LabVIEW Programming Environment
The LabVIEW graphical programming environment can be used to design and analyze a signal processing
system in a more time-ecient manner than with text-based programming environments. This chapter
provides an introduction to LabVIEW graphical programming. Also see [4], [5], and [2] to learn more about
LabVIEW graphical programming.
LabVIEW graphical programs are called virtual instruments (VIs). VIs run based on the concept of
dataow programming. This means that execution of a block or a graphical component is dependent on the
ow of data, or, more specically, a block executes after data is made available at all of its inputs. Block
output data are then sent to all other connected blocks. With dataow programming, one can perform
multiple operations in parallel because the execution of blocks is done by the ow of data and not by
sequential lines of code.
1.1.1 Virtual Instruments (VIs)

A VI consists of two major components: a front panel and block diagram. A front panel provides the user
interface of a program while a block diagram incorporates its graphical code. When a VI is located within
the block diagram of another VI, it is called a subVI. LabVIEW VIs are modular, meaning that one can run
any VI or subVI by itself.
1.1.1.1 Front Panel and Block Diagram

A front panel contains the user interfaces of a VI shown in a block diagram. VI inputs are represented by
controls such as knobs, pushbuttons and dials. VI outputs are represented by indicators such as graphs,
LEDs (light indicators) and meters. As a VI runs, its front panel provides a display or user interface of
controls (inputs) and indicators (outputs).
A block diagram contains terminal icons, nodes, wires and structures. Terminal icons, or interfaces
through which data are exchanged between a front panel and a block diagram, correspond to controls or
indicators that appear on a front panel. Whenever a control or indicator is placed on a front panel, a
terminal icon gets added to the corresponding block diagram. A node represents an object or block that
has input and/or output connectors and performs a certain function. SubVIs and functions are examples of
nodes. Wires establish the ow of data in a block diagram, and structures control the ow of data such as
repetitions or conditional executions. Figure 1.1 shows front panel and block diagram windows.

5
Figure 1.1: LabVIEW Windows: Front Panel and Block Diagram
1.1.1.2 Icon and Connector Pane

A VI icon is a graphical representation of a VI. It appears in the top right corner of a block diagram or a
front panel window. When a VI is inserted into a block diagram as a subVI, its icon is displayed.
A connector pane denes VI inputs (controls) and outputs (indicators). One can change the number of
inputs and outputs by using dierent connector pane patterns. In Figure 1.1, a VI icon is shown at the top
right corner of the block diagram, and its corresponding connector pane, with two inputs and one output, is
shown at the top right corner of the front panel.
1.1.2 Graphical Environment

1.1.2.1 Functions Palette
The Functions palette (see Figure 1.2) provides various function VIs or blocks to build a system. View this
palette by right-clicking on an open area of a block diagram. Note that this palette can be displayed only in
a block diagram.
Figure 1.2: Functions Palette
1.1.2.2 Controls Palette

The Controls palette (see Figure 1.3) features front panel controls and indicators. View this palette by
right-clicking on an open area of a front panel. Note that this palette can be displayed only in a front panel.
7
Figure 1.3: Controls Palette
1.1.2.3 Tools Palette

The Tools palette oers various mouse cursor operation modes for building or debugging a VI. The Tools
palette and the frequently used tools are shown in Figure 1.4.
Figure 1.4: Tools Palette
Each tool is used for a specic task. For example, use the wiring tool to wire objects in a block diagram.
If one enables the automatic tool selection mode by clicking on the Automatic Tool Selection button,
LabVIEW selects the best matching tool based on a current cursor position.
1.1.3 Building a Front Panel

In general, one constructs a VI by going back and forth between a front panel and block diagram, placing
inputs/outputs on the front panel and building blocks on the block diagram.
1.1.3.1 Controls
Controls make up the inputs to a VI. Controls grouped in the Numeric Controls palette(Controls →
Express → Num Ctrls) Buttons & Switches
are used for numerical inputs, controls grouped in the
palette(Controls → Express → Buttons) are used for Boolean inputs, and controls grouped in the
Text Controls palette(Controls →Express →Text Ctrls) are used for text and enumeration inputs.
These control options are displayed in Figure 1.5.
9
Figure 1.5: Control Palettes
1.1.3.2 Indicators
Indicators make up the outputs of a VI. Indicators grouped in the Numeric Indicators
palette(Controls→ Express → Numeric Inds) are used for numerical outputs, indicators grouped
in the LEDs palette(Controls → Express → LEDs) are used for Boolean outputs, indicators grouped
in the Text Indicators palette(Controls → Express → Text Inds) are used for text outputs, and
indicators grouped in the Graph Indicators palette(Controls → Express → Graph Indicators) are
used for graphical outputs. These indicator options are displayed in Figure 1.6.
Figure 1.6: Indicator Palettes
1.1.3.3 Align, Distribute and Resize Objects

The menu items on the front panel toolbar (see Figure 1.7) provide options to align and orderly distribute
objects on the front panel. Normally, after one places controls and indicators on a front panel, these options
can be used to tidy up their appearance.
11
Figure 1.7: Menu to Align, Distribute, Resize and Reorder Objects
1.1.4 Building a Block Diagram

1.1.4.1 Express VI and Function
Express VIs denote higher-level VIs congured to incorporate lower-level VIs or functions. These VIs are
displayed as expandable nodes with a blue background. Placing an Express VI in a block diagram opens
a conguration dialog window to adjust the Express VI parameters. As a result, Express VIs demand less
wiring. The conguration window can be opened by double-clicking on its Express VI.
Basic operations such as addition or subtraction are represented by functions. Figure 1.8 shows three
examples corresponding to three block diagram objects (VI, Express VI and function).
Figure 1.8: Block Diagram Objects: (a) VI, (b) Express VI, (c) Function
One can display subVIs or Express VIs as icons or expandable nodes. If a subVI is displayed as an
expandable node, the background appears yellow. Icons can be used to save space in a block diagram and
expandable nodes can be used to achieve easier wiring or better readability. One can resize expandable nodes
to show their connection nodes more clearly. Three appearances of a VI/Express VI are shown in Figure 1.9.
Figure 1.9: Icon versus Expandable Node
1.1.4.2 Terminal Icons

Front panel objects are displayed as terminal icons in a block diagram. A terminal icon exhibits an input
or output as well as its data type. Figure 1.10 shows two terminal icon examples consisting of a double
precision numerical control and indicator. As shown in this gure, one can display terminal icons as data
type terminal icons to conserve space in a block diagram.
Figure 1.10: Terminal Icon Examples Displayed in a Block Diagram

13
1.1.4.3 Wires
Wires transfer data from one node to another in a block diagram. Based on the data type of a data source,
the color and thickness of its connecting wires change.
Wires for the basic data types used in LabVIEW are shown in Figure 1.11. In addition to the data
types shown in this gure, there are some other specic data types. For example, the dynamic data type is
always used for Express VIs, and the waveform data type, which corresponds to the output from a waveform
generation VI, is a special cluster of waveform components incorporating trigger time, time interval and data
value.
Figure 1.11: Basic Wire Types
1.1.4.4 Structures
A structure is represented by a graphical enclosure. The graphical code enclosed in the structure gets
repeated or executed conditionally. A loop structure is equivalent to a for loop or a while loop statement in
text-based programming languages, while a case structure is equivalent to an if-else statement.
1.1.4.4.1 For Loop

A for loop structure is used to perform repetitions. As illustrated in Figure 1.12, the displayed border
indicates a for loop structure, where the count terminal represents the number of times the loop is to
be repeated. It is set by wiring a value from outside of the loop to it. The iteration terminal denotes the
number of completed iterations, which always starts at zero.
Figure 1.12: For Loop
1.1.4.4.2 While Loop

A while loop structure allows repetitions depending on a condition (see Figure 1.13). The conditional
terminal initiates a stop if the condition is true. Similar to a for loop, the iteration terminal
provides the number of completed iterations, always starting at zero.
Figure 1.13: While Loop
1.1.4.4.3 Case Structure

A case structure (see Figure 1.14) allows the running of dierent sets of operations depending on the value
it receives through its selector terminal, which is indicated by . In addition to Boolean type, the input
to a selector terminal can be of integer, string, or enumerated type. This input determines which case to
execute. The case selector shows the status being executed. Cases can be added or deleted as
needed.
Figure 1.14: Case Structure

15
1.1.5 Grouping Data: Array and Cluster

An array represents a group of elements having the same data type. An array consists of data elements
having a dimension up to 231 − 1. For example, if a random number is generated in a loop, it is appropriate
to build the output as an array because the length of the data element is xed at 1 and the data type is not
changed during iterations.
Similar to the structure data type in text-based programming languages, a cluster consists of a collection
of dierent data type elements. With clusters, one can reduce the number of wires on a block diagram by
bundling dierent data type elements together and passing them to only one terminal. One can add or
extract an individual element to or from a cluster by using the cluster functions such as Bundle by Name
and Unbundle by Name.
1.1.6 Debugging and Proling VIs

1.1.6.1 Probe Tool
VIs can be debugged as they run by checking values on wires with the Probe tool. Note that the Probe tool
can be accessed only in a block diagram window.
With the Probe tool, breakpoints and execution highlighting, one can identify the source of an incorrect
or an unexpected outcome. To visualize the ow of data during program execution, a breakpoint can be
used to pause the execution of a VI at a specic location.
1.1.6.2 Prole Tool

Timing and memory usage information in other words, how long a VI takes to run and how much memory
it consumes can be gathered with the Prole tool. It is required to make sure that a VI is stopped before
setting up a Prole window.
An eective way to become familiar with LabVIEW programming is to review examples. In the lab that
follows, we explore most of the key LabVIEW programming features by building simple VIs.
1.1.7 Containers and Decoration Tools

Containers and Decoration tools can be used to organize front panel controls and indicators. Container tools
are grouped in the Containers pallete(Controls → Modern → Containers → Classic → Classic
Containers) and Decoration tools are grouped in the Decorations pallete(Controls → Modern →
Decorations).
One can use Tab Control(Controls → Modern → Containers → Tab Control → Classic →
Classic Containers →Tab Control) to display various controls and indicators within a limited screen area.
This feature helps one to organize controls and indicators under dierent tabs as illustrated in Figure 1.15.
To add more tabs or delete tabs, right-click the border area and choose one of the following options: Add
Page After, Add Page Before, Duplicate Page or Remove Page.
Figure 1.15: Tab Control

17
2
1.2 Lab 1: Introduction to LabVIEW
The objective of this lab is to oer an initial hands-on experience in building a VI. More detailed explanations
of the LabVIEW features mentioned here can be found in the [4], [5], and [2]. One can launch LabVIEW
2011 (the latest version at the time of this publication) by double-clicking on the LabVIEW 2011 icon, which
opens the dialog window shown in Figure 1.16.
Figure 1.16: Starting LabVIEW
1.2.1 Building a Simple VI

To become familiar with the LabVIEW programming environment, let us calculate the sum and average of
two input values in the following step-by-step example.
1.2.1.1 Sum and Average VI Example Using Graphical Programming

To create a new VI, click on the Blank VI under New, as shown in Figure 1.17. This can also be done by
choosing File → New VI from the menu. As a result, a blank front panel and a blank block diagram
window appear, see Figure 1.17. Remember that a front panel and block diagram coexist when one builds
a VI, meaning that every VI will have both a front panel and an associated block diagram.

Figure 1.17: Blank VI
The number of VI inputs and outputs is dependent on the VI function. In this example, two inputs
and two outputs are needed, one output generating the sum and the other generating the average of two
input values. Create the inputs by locating two numeric controls on the front panel. This can be done by
right-clicking on an open area of the front panel to bring up the Controls palette, followed by choosing
Controls → Modern → Numeric → Numeric Control. Each numeric control automatically places
a corresponding terminal icon on the block diagram. Double-clicking on a numeric control highlights its
counterpart on the block diagram and vice versa.
Tools Palette, which can be
Next, label the two inputs as x and y using the Labeling tool from the
displayed by choosing View → Tools Palette from the menu bar. Choose the Labeling tool and click on
the default labels, Numeric and Numeric 2, to edit them. Alternatively, if the automatic tool selection
mode is enabled by clicking Automatic Tool Selection in the Tools Palette, the labels can be edited
by simply double-clicking on the default labels. Editing a label on the front panel changes its corresponding
terminal icon label on the block diagram and vice versa.
Similarly, the outputs are created by locating two numeric indicators (Controls → Modern → Nu-
meric →Numeric Indicator) on the front panel. Each numeric indicator automatically places a corre-
sponding terminal icon on the block diagram. Edit the labels of the indicators to read Sum and Average.
For a better visual appearance, one can align, distribute and resize objects on a front panel window using
the front panel toolbar. To do this, select the objects to be aligned or distributed and apply the appropriate
option from the toolbar menu. Figure 1.18 shows the conguration of the front panel just created.
19
Figure 1.18: Front Panel Conguration
Now build a graphical code on the block diagram to perform the summation and averaging operations.
Note that <Ctrl-E> toggles between a front panel and a block diagram window. If objects on a block diagram
are too close to insert other functions or VIs in-between, one can insert a horizontal or vertical space by
holding down the <Ctrl> key to create space horizontally and/or vertically. As an example, Figure 1.19b
illustrates a horizontal space inserted between the objects shown in Figure 1.18a.
Figure 1.19: Inserting Horizontal/Vertical Space: (a) Creating Space While Holding Down the <Ctrl>
Key, (b) Inserted Horizontal Space.
Next, place an Add function (Functions →Express →Arith & Compar →Express Numeric
→Add) and a Divide function (Functions →Express →Arith & Comp →Express Numeric
→Divide) on the block diagram. Enter the divisor, in this case 2, in a Numeric Constant(Functions
→Express →Arith & Compar →Express Numeric →Numeric Constant) and connect it to the y
terminal of the Divide function using the Wiring tool.
To achieve proper data ow, wire functions, structures and terminal icons on a block diagram using the
Wiring tool. To wire these objects, point the Wiring tool at the terminal of the function or subVI to be wired,
left-click on the terminal, drag the mouse to a destination terminal and left-click once again. Figure 1.20
illustrates the wires placed between the terminals of the numeric controls and the input terminals of the
Add function. Notice that the label of a terminal gets displayed whenever one moves the cursor over the
terminal if the automatic tool selection mode is enabled. Also, note that the Run button on the toolbar
remains broken until one completes the wiring process.
21
Figure 1.20: Wiring Block Diagram Objects.
For better block diagram readability, one can clean up wires hidden behind objects or crossed over other
wires by right-clicking on them and choosing Clean Up Wire from the shortcut menu. Any broken wires
can be cleared by pressing <Ctrl-B> or Edit →Remove Broken Wires.
To view or hide the label of a block diagram object, such as a function, right-click on the object and
check (or uncheck) Visible Items →Label from the shortcut menu. Also, one can show a terminal icon
corresponding to a numeric control or indicator as a data type terminal icon by right-clicking on the terminal
icon and unchecking View As Icon from the shortcut menu. Figure 1.21 shows an example where the
numeric controls and indicators are depicted as data type terminal icons. The notation DBL indicates
double precision data type.
Figure 1.21: Completed Block Diagram.
It is worth noting that there is a shortcut to build the above VI. Instead of choosing the numeric
controls, indicators or constants from the Controls or Functions palette, one can use the shortcut menu
Create, activated by right-clicking on a terminal of a block diagram object such as a function or a subVI.
As an example of this approach, create a blank VI and locate an Add function. Right-click on its x terminal
and choose Create →Control from the shortcut menu to create and wire a numeric control or input. This
locates a numeric control on the front panel as well as a corresponding terminal icon on the block diagram.
The label is automatically set to x. Create a second numeric control by right-clicking on the y terminal
Add function. Next, right-click on the output terminal of the Add function and choose Create
of the
→Indicator from the shortcut menu. A data type terminal icon, labeled as x+y, is created on the block
diagram as well as a corresponding numeric indicator on the front panel.
Next, right-click on the y terminal of the Divide function to choose Create →Constant from the
shortcut menu. This creates a numeric constant as the divisor and wires its y terminal. Type the value 2
in the numeric constant. Right-click on the output terminal of the Divide function, labeled as x/y, and
choose Create →Indicator from the shortcut menu. If the wrong option is chosen, the terminal does not
get wired. An incorrect terminal option can easily be changed by right-clicking on the terminal and choosing
Change to Control from the shortcut menu.
To save the created VI for later use, choose File →Save from the menu or press <Ctrl-S> to bring up
a dialog window to enter a name. Type Sum and Average as the VI name and click Save.
To test the functionality of the VI, enter some sample values in the numeric controls on the front panel
23
and run the VI by choosing Operate →Run, by pressing <Ctrl-R> or by clicking the Run button on
the toolbar. From the displayed output values in the numeric indicators, the functionality of the VI can be
veried. Figure 1.22 illustrates the outcome after running the VI with two inputs, 10 and 15.
Figure 1.22: VI Verication
1.2.2 SubVI Creation

If it is desired to use a VI as part of a higher-level VI, one needs to congure its connector pane. A connector
pane assigns inputs and outputs of a subVI to its terminals through which data are exchanged.
The default pattern of a connector pane is determined based on the number of controls and indicators.
In general, the terminals on the left side of a connector pane pattern are used for inputs and the ones on the
right side for outputs. One can add terminals to or remove them from a connector pane by right-clicking and
choosing Add Terminal or Remove Terminal from the shortcut menu. If the number of inputs/outputs
or the distribution of terminals are changed, the connector pane pattern can be replaced with a new one
by right-clicking and choosing Patterns from the shortcut menu. Once a pattern is selected, one needs to
reassign each terminal to a control or an indicator by using the Wiring tool or by enabling the automatic
tool selection mode.
Figure 1.23a illustrates how to assign a Sum and Average VI terminal to a numeric control. The completed
connector pane is shown in Figure 1.23b. Notice that the output terminals have thicker borders. The color
of a terminal reects its data type.
Figure 1.23: Connector Pane: (a) Assigning a Terminal to a Control, (b) Completed Terminal Assign-
ment.
Considering that a subVI icon is displayed on the block diagram of a higher-level VI, it is important to
edit the subVI icon for it to be explicitly identiable. Double-clicking on the top-right corner icon of a block
diagram opens the Icon Editor. The Icon Editor tools are similar to those in other graphical editors, such
as Microsoft Paint. Editing the Sum and Average VI icon is illustrated in Figure 1.24.
Figure 1.24: Editing SubVI Icon.

25
A subVI can also be created from a section of a VI. To do so, select the nodes on the block diagram to
be included in the subVI, as shown in Figure 1.25a. Then, choose Edit →Create SubVI to insert a new
subVI icon. Figure 1.25b illustrates the block diagram with an inserted subVI. One can open and edit this
subVI by double-clicking on its icon on the block diagram. Save this subVI as Sum and Average.vi. This
subVI performs the same function as the original Sum and Average VI.
Figure 1.25: Creating a SubVI: (a) Selecting Nodes to Make a SubVI, (b) Inserted SubVI Icon.
1.2.3 Using Structures and SubVIs

Now let us consider another example to understand the use of structures and subVIs. In this example, we
use a VI to show the sum and average of two input values, which are altered in a continuous fashion. If the
average of the two inputs becomes greater than a preset threshold value, a LED warning light turns on.
First, build a front panel as shown in Figure 1.26a. For the inputs, consider two Knobs(Controls
→Modern →Numeric →Knob). Adjust the size of the knobs by using the Positioning tool. One can
modify knob properties such as precision and data type by right-clicking and choosing Properties from the
shortcut menu. A Knob Properties dialog box opens and an Appearance tab is shown by default. Edit
the label of one of the knobs to read Input 1. Select the Data Type tab, click Representation and select
Byte to change the data type from double precision to byte. One can also perform this by right-clicking on
the knob and choosing Representation →Byte from the shortcut menu. In the Data Type tab, a default
value needs to be specied. In this example, the default value is considered to be 0. The default value can
be set by right-clicking on the control and choosing Data Operations →Make Current Value Default
Data
from the shortcut menu. Also, this control can be set to a default value by right-clicking and choosing
Operations →Reinitialize to Default Value from the shortcut menu.
Label the second knob as Input 2 and repeat all the adjustments as carried out for the rst knob except for
the data representation part. Specify the data type of the second knob to be double precision to demonstrate
the dierence in the outcome. As the nal front panel conguration step, align and distribute the objects
using the appropriate buttons on the front panel toolbar.
To set the outputs, locate and place a numeric indicator, a round LED (Controls →Modern →Boolean
→Round LED) and a gauge (Controls →Modern →Numeric →Gauge). Edit the labels of the indi-
cators as shown in Figure 1.26.
Figure 1.26: Example of Structure and SubVI: (a) Front Panel, (b) Block Diagram.
Locate a Greater or Equal? function from Functions →Programming →Comparison →Greater

or Equal? to compare the average output of the subVI with a threshold value. Create a wire branch on
the wire between the Average terminal of the subVI and its indicator via the Wiring tool. Then, extend
this wire to the x terminal of the Greater or Equal? function. Right-click on the y terminal of the Greater
or Equal? function and choose Create →Constant to place a numeric constant. Enter 9 in the numeric
constant and wire the round LED, labeled as Warning, to the x>=y? terminal of this function to provide a
Boolean value.
To run the VI continuously, use a while loop structure. Choose Functions →Programming
→Structures →While Loopto create a while loop. Change the size by dragging the mouse to enclose
the objects in the while loop, as illustrated in Figure 1.27.
27
Figure 1.27: While Loop Enclosure.
Once this structure is created, its boundary, together with the loop iteration terminal and condi-
tional terminal , get shown on the block diagram. If one creates the while loop by using Functions
→Programming →Structures → While Loop, the Stop button is not included as part of the structure.
One can create this button by right-clicking on the conditional terminal and choosing Create →Control
from the shortcut menu. It is possible to wire a Boolean condition to a conditional terminal, instead of a
Stop button, to stop the loop programmatically.
Next run the VI to verify its functionality. After clicking the Run button on the toolbar, adjust the
knobs to alter the inputs. Verify whether the average and sum are displayed correctly in the gauge and
numeric indicators. Note that only integer values can be entered via the Input 1 knob while real values can
be entered via the Input 2 knob. This is due to the data types associated with these knobs. The Input 1
knob is set to byte type, in other words, I8 or 8-bit signed integer. As a result, one can enter only integer
values within the range -128 and 127. Considering that the minimum and maximum values of this knob are
set to 0 and 10, respectively, one can enter only integer values from 0 to 10 for this input.
Figure 1.28: Front Panel as VI Runs.
1.2.4 Debugging VIs: Probe Tool

Use the Probe tool to observe data that are being passed while a VI is running. A probe can be placed on
a wire by using the Probe tool or by right-clicking on a wire and choosing Probe from the shortcut menu.
Probes can also be placed while a VI is running.
Placing probes on wires creates probe windows through which one can observe intermediate values. As
an example of using custom probes, use four probe windows at the probe locations 1 through 4 in the Sum
and Average VI to probe the values at those locations. These probes and their locations are illustrated in
Figure 1.29.
29
Figure 1.29: Probe Tool.
1.2.5 Prole Tool

With the Prole tool, one can gather timing and memory usage information. Make sure to stop the VI
before selecting Tools →Prole →Performance and Memory to open a Prole window.
Place a checkmark in the Timing Statistics checkbox to display timing statistics of the VI. The Timing
Details option oers more detailed VI statistics such as drawing time. To prole memory usage as well as
timing, check the Memory Usage checkbox after checking the Prole Memory Usage checkbox. Note
that this option can slow down VI execution. Start proling by clicking the Start button on the proler,
then run the VI. Obtain a snapshot of the proler information by clicking on the Snapshot button. After
viewing the timing information, click the Stop button. The prole statistics can be stored in a text le by
clicking the Save button.
An outcome of the proler is shown in Figure 1.30 after running the Sum and Average or L1.1 VI. [4]
provides more details on the Prole tool.
Figure 1.30: Prole Window after Running Sum and Average VI.
1.2.6 Lab Exercises

Exercise 1.2.1 (Solution on p. 31.)
Build a VI to compute the variance of an array x. The variance σ is dened as:
N
1 X 2
σ= (xj − µ) (1.1)
N j=1
where µdenotes the average of the array x. For x, use all the integers from 1 to 1000.

Build a VI to check whether a given positive integer n is a prime number and display a warning
message if it is not a prime number.

Build a VI to generate the rst N prime numbers and store them using an indexing array. Display
the outcome.

Build a VI to sort N integer numbers (positive or negative) in ascending or descending order.
31
Solutions to Exercises in Chapter 1
Solutions to Lab 1: Introduction to LabVIEW

Solution to Exercise 1.2.1 (p. 30)
Insert Solution Text Here
Chapter 2
LabVIEW MathScript and Hybrid

Programming
33
34 CHAPTER 2. LABVIEW MATHSCRIPT AND HYBRID PROGRAMMING
1
2.1 LabVIEW MathScript and Hybrid Programming
In signals and systems lab courses, .m le coding is widely used. LabVIEW MathScript is a feature of the
newer versions of LabVIEW that allows one to include .m les within its graphical environment. As a result,
one can perform hybrid programming, that is, a combination of textual and graphical programming, when
using this feature. This chapter provides an introduction to MathScript or .m le textual coding. See [6]
and [7] for advanced MathScript aspects.
MathScripting can be done via the LabVIEW MathScript interactive window or node. The LabVIEW
MathScript interactive window, shown in Figure 2.1, consists of a Command Window, an Output Window
and a MathScript Window. The Command Window interface allows one to enter commands and debug
script or to view help statements for built-in functions. The Output Window is used to view output values
and the MathScript Window interface to display variables and command history as well as edit scripts. With
script editing, one can execute a group of commands or textual statements.

35
Figure 2.1: LabVIEW MathScript Interactive Window
A LabVIEW MathScript node represents the textual .m le code via a blue rectangle as shown in
Figure 2.2. Its inputs and outputs are dened on the border of this rectangle for transferring data between
the graphical environment and the textual code. For example, as indicated in Figure 2.2, the input variables
on the left side, namely lf, hf and order, transfer values to the .m le script, and the output variables on
the right side, F and sH, transfer values to the graphical environment. This process allows .m le script
variables to be used within the LabVIEW graphical programming environment.
Figure 2.2: LabVIEW MathScript Node Interface

37
2
2.2 Lab 2: LabVIEW MathScript and Hybrid Programming
2.2.1 Arithmetic Operations

There are four basic arithmetic operators in .m les:
+ addition
- subtraction
* multiplication
/ division (for matrices, it also means inversion)
The following three operators work on an element-by-element basis:
.* multiplication of two vectors, element-wise
./ division of two vectors, element-wise
.^ raising all the elements of a vector √to a power
3
As an example, to evaluate the expression a + bd − 4c , where a = 1.2, b = 2.3, c = 4.5and d = 4, type
the following commands in the Command Window to get the answer (ans) :
a=1.2;
b=2.3;
c=4.5;
d=4;
a^3+sqrt(b*d)-4*c
ans =
-13.2388
Note the semicolon after each variable assignment. If the semicolon is omitted, the interpreter echoes
back the variable value.
2.2.2 Vector Operations

Consider the vectors x = [x1 , x2 , ..., xn ]and y = [y1 , y2 , ..., yn ]. The following operations indicate the resulting
vectors:
x*.y = h[x1 y1 , x2 y2 , ...,ixn yn ]
x./y = xy , xy , ..., xy
1
1
2
3
n
n
x.^p = [xp1 , xp2 , ..., xpn ]

Note that because the boldfacing of vectors/matrices are not used in .m les, in the notation adopted in
this book, no boldfacing of vectors/matrices is shown to retain consistency with .m les.
The arithmetic operators + and can be used to add or subtract matrices, vectors or scalars. Vectors
denote one-dimensional arrays and matrices denote multidimensional arrays. For example,
x=[1,3,4]
y=[4,5,6]
x+y
ans=
5 8 10
In this example, the operator + adds the elements of the vectors x and y, element by element, assuming
that the two vectors have the same dimension, in this case 1×3 or one row with three columns. An error
occurs if one attempts to add vectors having dierent dimensions. The same applies for matrices.
P
To compute the dot product of two vectors (in other words, i xi yi ), use the multiplication operator
`*' as follows:
x*y'
ans =
43
Note the single quote after y denotes the transpose of a vector or a matrix.

To compute an element-by-element multiplication of two vectors (or two arrays), use the following oper-
ator:
x .* y
ans =
4 15 24 h
That is, x .* y means [1 × 4, 3 × 5, 4 × 6] = 4 15 24 .
2.2.3 Complex Numbers

LabVIEW MathScript supports complex numbers. The imaginary number is denoted with the symbol i or
j, assuming that these symbols have not been used any other place in the program. It is critical to avoid
such a symbol conict for obtaining correct outcome. Enter the following and observe the outcomes:
z=3 + 4i % note the multiplication sign `*' is not needed after 4
conj(z) % computes the conjugate of z
angle(z) % computes the phase of z
real(z) % computes the real part of z
imag(z) % computes the imaginary part of z
abs(z) % computes the magnitude of z
One can also dene an imaginary number with any other user-specied variables. For example, try the
following:
img=sqrt(-1)
z=3+4*img
exp(pi*img)
2.2.4 Array Indexing

In .m les, all arrays (vectors) are indexed starting from 1 − in other words, x(1) denotes the rst element
of the array x. Note that the arrays are indexed using parentheses (.) and not square brackets [.], as done
in C/C++. To create an array featuring the integers 1 through 6 as elements, enter:
x=[1,2,3,4,5,6]
Alternatively, use the notation `:'
x=1:6
This notation creates a vector starting from 1 to 6, in steps of 1. If a vector from 1 to 6 in steps of 2 is
desired, then type:
x=1:2:6
ans =
1 3 5
Also, examine the following code:
ii=2:4:17
jj=20:-2:0
ii=2:(1/10):4
One can easily extract numbers in a vector. To concatenate an array, the example below shows how to
use the operator `[ ]':
x=[1:3 4 6 100:110]
To access a subset of this array, try the following:
x(3:7)
length(x) % gives the size of the array or vector
x(2:2:length(x))
39
2.2.5 Allocating Memory

One can allocate memory for one-dimensional arrays (vectors) using the command zeros. The following
command allocates memory for a 100-dimensional array:
y=zeros(100,1);
y(30)
ans =
0
One can allocate memory for two-dimensional arrays (matrices) in a similar fashion. The command
y=zeros(4,5)
denes a 4 by 5 matrix. Similar to the command zeros, the command ones can be used to dene a vector
containing all ones,
y=ones(1,5)
ans=
1 1 1 1 1
2.2.6 Special Characters and Functions

Some common special characters used in .m les are listed below for later reference:
Symbol Meaning
pi π (3.14.....)
^ indicates power (for example, 3^2=9)
NaN not-a-number, obtained when encountering unde-

ned operations, such as 0/0
Inf Represents +∞
; indicates the end of a row in a matrix; also used to
suppress printing on the screen (echo o )
% comments − anything to the right of % is ignored

by the .m le interpreter and is considered to be
comments
` denotes transpose of a vector or a matrix; also used

to dene strings, for example, str1='DSP'
... denotes continuation; three or more periods at the

end of a line continue current function to next line
Table 2.1: Some common special characters used in .m les
Some special functions are listed below for later reference:
Function Meaning
sqrt indicates square root, for example, sqrt(4)=2
abs absolute value | . |, for example, abs(-3)=3
length length(x) gives the dimension of the array x
sum nds sum of the elements of a vector
nd nds indices of nonzero

Table 2.2: Some common functions used in .m les
Here is an example of the function length,

x=1:10;
length(x)
ans =
10
The function nd returns the indices of a vector that are non-zero. For example,
I = find(x>4) nds all the indices of x greater than 4. Thus, for the above example:
find(x> 4)
ans =
5 6 7 8 9 10
2.2.7 Control Flow

.m les have the following control ow constructs:
• if statements
• switch statements
• for loops
• while loops
• break statements
The constructs if, for, switch and while need to terminate with an end statement. Examples are provided
below:
if
x=-3;
if x>0
str='positive'
elseif x<0
str='negative'
elseif x== 0
str='zero'
else
str='error'
end
See the value of 'str' after executing the above code.
while
x=-10;
while x<0
x=x+1;
end
See the value of x after executing the above code.
for loop
x=0;
for j=1:10
x=x+j;
end
The above code computes the sum of all the numbers from 1 to 10.
break
With the break statement, one can exit early from a for or a while loop:
x=-10;
while x<0
x=x+2;
41
if x = = -2
break;
end
end
LabVIEW MathScript supports the relational and logical operators listed below.
Relational Operators
Symbol Meaning
<= less than equal
< less than
>= greater than equal
> greater than
== equal
∼= not equal
Table 2.3: Relational Operators
Logical Operators
Symbol Meaning
& AND
| OR
∼ NOT
Table 2.4: Logical Operators
2.2.8 Programming in the LabVIEW MathScript Window

The MathScript feature allows one to include .m les, which can be created using any text editor. To activate
the LabVIEW MathScript interactive window, select Tools → MathScript Window from the main menu.
To open the LabVIEW MathScript text editor, click the Script tab of the LabVIEW MathScript Window
(see Figure 2.3). After typing the .m le textual code, save it and click on the Run script button (green
arrow) to run it.
For instance, to write a program to compute the average (mean) of a vector x, the program should use as
its input the vector x and return the average value. To write this program, follow the steps outlined below.
Type the following in the empty script:
x=1:10
L=length(x);
sum=0;
for j=1:L
sum=sum+x(j);
end
y=sum/L % the average of x
From the Editor pull-down menu, go to File → Save Script As and enter average.m for the le name.
Then click on the Run script button to run the program. Figure 2.3 shows the LabVIEW MathScript
interactive window after running the program.
Figure 2.3: LabVIEW MathScript Interactive Window after Running the Program Average
2.2.9 Sound Generation

Assuming the computer used has a sound card, one can use the function sound to play back speech or audio
les through its speakers. That is, sound(y,FS) sends the signal in a vector y (with sample frequency FS) out
to the speaker. Stereo sounds are played on platforms that support them, with y being an N-by-2 matrix.
Try the following code and listen to a 400 Hz tone:
t=0:1/8000:1;
x=cos(2*pi*400*t);
sound(x,8000);
Now generate a noise signal by typing:
noise=randn(1,8000); % generate 8000 samples of noise
sound(noise,8000);
The function randn generates Gaussian noise with zero mean and unit variance.
43
2.2.10 Loading and Saving Data

One can load or store data using the commands load and save. To save the vector x of the above code in
the le data.mat, type:
save data x
Note that LabVIEW MathScript data les have the extension .mat. To retrieve the data saved, type:
load data
The vector x gets loaded in memory. To see memory contents, use the command whos,
whos
Variable Dimension Type x 1x8000 double array
The command whos gives a list of all the variables currently in memory, along with their dimensions. In
the above example, x contains 8000 samples.
To clear up memory after loading a le, type clear all when done. This is important because if one
does not clear all the variables, one could experience conicts with other programs using the same variables.
2.2.11 Reading Wave and Image Files

With LabVIEW MathScript, one can read data from dierent le types (such as .wav, .jpeg and .bmp) and
load them in a vector.
To read an audio data le with .wav extension, use the following command:
[y Fs]=wavread(`filename')
This command reads a wave le specied by the string lename and returns the sampled data in y with
the sampling rate of Fs (in hertz).
To read an image le, use the following command:
[y]=imread(`filename', `filetype')
This command reads a grayscale or color image from the string lename, where letype species the
format of the le and returns the image data in the array y.
2.2.12 Signal Display

Several tools are available in LabVIEW to display data in a graphical format. Throughout the book, signals
in both the time and frequency domains are displayed using the following two graph tools.
Waveform GraphDisplays data acquired at a constant rate.
XY GraphDisplays data acquired at a non-constant rate, such as data acquired when a trigger occurs.
A waveform graph can be created on a front panel by choosing Controls→ Express → Waveform Graph.
Figure 2.4 shows a waveform graph and the waveform graph elements which can be opened by right-clicking
on the graph and selecting Visible Items from the shortcut menu.
Figure 2.4: Waveform Graph
Often a waveform graph is tied with the function Build Waveform(Function→ Programming →
Waveform → Build Waveform) to calibrate the x scale (which is time scale for signals), as shown in
Figure 2.5.
Figure 2.5: Build Waveform Function and Waveform Graph
Create an XY graph from a front panel by choosing Controls→ Express → XY Graph. Figure 2.6
shows an XY graph and its dierent elements.
45
Figure 2.6: XY Graph
An XY graph displays a signal at a non-constant rate, and one can tie together its X and Y vectors to
Build XY Graph function. This function automatically appears on the block
display the signal via the
diagram when placing an XY graph on the front panel, as shown in Figure 2.7. Note that one can use the
function Bundle (Functions → Programming → Cluster & Variant → Bundle) instead of Build XY
Graph.
Figure 2.7: Build XY Graph Function

2.2.13 Hybrid Programming

As stated earlier, the LabVIEW MathScript feature can be used to perform hybrid programming, in other
words, a combination of textual .m les and graphical objects. Normally, it is easier to carry out math
operations via .m les while maintaining user interfacing, interactivity and analysis in the more intuitive
graphical environment of LabVIEW. Textual .m le codes can be typed in or copied and pasted into Lab-
VIEW MathScript nodes.
2.2.13.1 Sum and Average VI Example Using Hybrid Programming

Sum and Average VI Example Using Hybrid Programming
Choose Functions →Programming →Structures → MathScript to create a LabVIEW MathScript
node (see Figure 2.8). Change the size of the window by dragging the mouse.
Figure 2.8: LabVIEW MathScript Node Creation
Now build the same program average using a LabVIEW MathScript node. The inputs to this program
consist of x and y. To add these inputs, right-click on the border of the LabVIEW MathScript node and
click on the Add Input option (see Figure 2.9).
47
Figure 2.9: (a) Adding Inputs, (b) Creating Controls
After adding these inputs, create controls to change the inputs interactively via the front panel. By
right-clicking on the border, add outputs in a similar manner. An important issue to consider is the selection
of output data type. The outputs of the Sum and Average VI are scalar quantities. Choose data types by
right-clicking on an output and selecting the Choose Data Type option (see Figure 2.10).
Figure 2.10: (a) Adding Outputs, (b) Choosing Data Types

49
Finally, add numeric indicators in a similar fashion as indicated earlier. Figure 2.11 shows the completed
block diagram and front panel.
Figure 2.11: (a) Completed Block Diagram, (b) Completed Front Panel
2.2.13.2 Building a Signal Generation System Using Hybrid Programming

In this section, let us see how to generate and display aperiodic continuous-time signals or pulses in the time
domain. One can represent such signals with a function of time. For simulation purposes, a representation
of time tis needed. Note that the time scale is continuous while computer programs operate in a discrete
fashion. This simulation can be achieved by considering a very small time interval. For example, if a 1-second
duration signal in millisecond increments (time interval of 0.001 second) is considered, then one sample every
1 millisecond and a total of 1000 samples are generated for the entire signal. This continuous-time signal
approximation is discussed further in later chapters. It is important to note that there is a nite number of
samples for a continuous-time signal, and, to dierentiate this signal from a discrete-time signal, one must
assign a much higher number of samples per second (very small time interval).
Figure 2.12: Continuous-Time Signals
Figure 2.12 shows two continuous-time signals x1 (t) and x2 (t)with a duration of 3 seconds. By setting
the time interval dt to 0.001 second, there is a total of 3000 samples at t = 0, 0.001, 0.002, 0.003, ......., 2.999
seconds.
The signal x1 (t) can be represented mathematically as follows:
0 0≤t<1
x1 (t) = { 1 1≤t<2 (2.1)
0 2≤t<3
To simulate this signal, use the LabVIEW MathScript functions ones and zeros. The signal value is zero
during the rst second, which means the rst 1000 samples are zero. This portion of the signal is simulated
with the function zeros(1,1000). In the next second (next 1000 samples), the signal value is 2, and this
portion is simulated by the function 2*ones(1,1000). Finally, the third portion of the signal is simulated by
the function zeros(1,1000). In other words, the entire duration of the signal is simulated by the following
.m le function:
x1=[ zeros(1,1/dt) 2*ones(1,1/dt) zeros(1,1/dt)]
The signal x2 (t) can be represented mathematically as follows:
2t 0≤t<1
x2 (t) = { −2t + 4 1≤t<2 (2.2)
0 2≤t<3
Use a linearly increasing or decreasing vector to represent the linear portions. The time vectors for the
three portions or segments of the signal are 0:dt:1-dt, 1:dt:2-dt and 2:dt:3-dt. The rst segment is
a linear function corresponding to a time vector with a slope of 2; the second segment is a linear function
corresponding to a time vector with a slope of -2 and an oset of 4; and the third segment is simply a
constant vector of zeros. In other words, simulate the entire duration of the signal for any value of dt by the
following .m le function:
x2=[2*(0:dt:(1-dt)) -2*(1:dt:(2-dt))+4 zeros(1,1/dt)].
Figure 2.13 and Figure 2.14 show the block diagram and front panel of the above signal generation system,
respectively. Display the signals using a Waveform Graph(Controls→ Express → Waveform Graph)
and a Build Waveform function (Function→ Programming → Waveform → Build Waveform).
Note that the default data type in MathScript is double precision scalar. So whenever an output possesses
51
any other data type, one needs to right-click on the output and select the Choose Data Type option. In
this example, x1 and x2 are double precision one-dimensional arrays that are specied accordingly.
Figure 2.13: Block Diagram of a Signal Generation System

Figure 2.14: Front Panel of a Signal Generation System
2.2.13.3 Building a Periodic Signal Generation System Using Hybrid Programming

In this section, build a simple periodic signal generation system in hybrid mode to set the stage for the
chapters that follow. This system involves generating a periodic signal in textual mode and displaying it in
graphical mode. Modify the shape of the signal (sine, square, triangle or sawtooth) as well as its frequency
and amplitude by using appropriate front panel controls. The block diagram and front panel of this system
53
using a LabVIEW MathScript node are shown in Figure 2.15 and Figure 2.16, respectively. The front panel
includes the following three controls:
Waveform type Select the shape of the input waveform as either sine, square, triangular or sawtooth
waves.
Amplitude Control the amplitude of the input waveform.
Frequency Control the frequency of the input waveform.
Figure 2.15: Periodic Signal Generation System Block Diagram

Figure 2.16: Periodic Signal Generation System Front Panel
To build the block diagram, rst write a .m le code to generate four types of waveforms using the .m
le functions sin, square and sawtooth. To change the amplitude and frequency of the waveforms, use two
controls named Amplitude (A) and Frequency (f ). Waveform Type (w) is another input controlled by the
Enum Control for selecting the waveform type. With this control, one can select from multiple inputs.
Create an Enum Control from the front panel by invoking Controls → Modern → Ring & Enum →
Enum. Right-click on the Enum Control to select properties and the edit item tab to choose dierent
items as shown in Figure 2.17. After inserting each item, the digital display shows the corresponding number
Enum Control.
value for that item, which is the output of the
Waveform Graph(Controls→ Express → Waveform Graph)
Finally, display the waveforms with a
and a Build Waveform function (Function→ Programming → Waveform → Build Waveform).
55
Figure 2.17: Enum Control Properties
2.2.14 Lab Exercises

Write a .m le code to add all the numbers corresponding to the even indices of an array. For
instance, if the array x is specied as x = [1, 3, 5, 10], then 13 (= 3+10) should be returned. Use
the program to nd the sum of all even integers from 1 to 1000. Run your code using the LabVIEW
MathScript interactive window. Also, redo the code where x is the input vector and y is the sum
of all the numbers corresponding to the even indices of x.

2. Explain what the following .m le does:
L=length(x);
for j=1:L
if x(j) < 0
x(j)=-x(j);
end
end
Rewrite this program without using a for loop.

3. Write a .m le code that implements the following hard-limiting function:
0.2 t ≥ 0.2
x (t) = { (2.3)
−0.2 t < 0.2
For t, use 1000 random numbers generated via the function rand.

4. Build a hybrid VI to generate two sinusoid signals with the frequencies f1 Hz and f2 Hz and the
amplitudes A1 and A2, based on a sampling frequency of 8000 Hz with the number of samples being
256. Set the frequency ranges from 100 to 400 Hz and set the amplitude ranges from 20 to 200.
Generate a third signal with the frequency f3 = (mod (lcm (f1, f2), 400) + 100) Hz, where mod and
lcm denote the modulus and least common multiple operation, respectively, and the amplitude A3
is the sum of the amplitudes A1 and A2. Use the same sampling frequency and number of samples
as specied for the rst two signals. Display all the signals using the legend on the same waveform
graph and label them accordingly.
57
Solutions to Lab 2: LabVIEW MathScript and Hybrid Programming

Chapter 3
Convolution and Linear Time-Invariant

Systems
1
3.1 Convolution and Linear Time-Invariant Systems
3.1.1 Convolution and Its Numerical Approximation

The output y (t) of a continuous-time linear time-invariant (LTI) system is related to its input x (t) and the
system impulse response h (t) through the convolution integral expressed as (for details on the theory of
convolution and LTI systems, refer to signals and systems textbooks, for example, references [9] - [15] ):
Z∞
y (t) = h (t − τ ) x (τ ) dτ (3.1)
−∞
For a computer program to perform the above continuous-time convolution integral, a numerical approx-
imation of the integral is needed noting that computer programs operate in a discrete not continuous
fashion. One way to approximate the continuous functions in the Equation (1) integral is to use piecewise
constant functions. Dene δ∆ (t) to be a rectangular pulse of width ∆ and height 1, centered at t = 0:
1 −∆/2 ≤ t ≤ ∆/2
δ∆ (t) = { (3.2)
0 otherwise
Approximate a continuous function x (t) with a piecewise constant function x∆ (t) as a sequence of pulses
spaced every ∆ seconds in time with heights x (k∆):
∞
X
x∆ (t) = x (k∆) δ∆ (t − k∆) (3.3)
k=−∞
It can be shown in the limit as ∆ → 0, x∆ (t) → x (t). As an example, Figure 3.1 shows the approximation
of a decaying exponential x (t) = exp − 2t starting from 0 using ∆ = 1. Similarly, h (t) can be approximated
by
∞
X
h∆ (t) = h (k∆) δ∆ (t − k∆) (3.4)
k=−∞
59
60 CHAPTER 3. CONVOLUTION AND LINEAR TIME-INVARIANT SYSTEMS
One can thus approximate the convolution integral by convolving the two piecewise constant signals as
follows:
Z∞
y∆ (t) = h∆ (t − τ ) x∆ (τ ) dτ (3.5)
−∞
Figure 3.1: Approximation of a Decaying Exponential with Rectangular Strips of Width 1
Notice that y∆ (t) is not necessarily a piecewise constant. For computer representation purposes, discrete
output values are needed, which can be obtained by further approximating the convolution integral as
indicated below:
∞
X
y∆ (n∆) = ∆ x (k∆) h ((n − k) ∆) (3.6)
k=−∞
If one represents the signals h∆ (t) and x∆ (t) in a .m le by vectors containing the values of the signals at
t = n∆, then Equation (5) can be used to compute an approximation to the convolution of
P∞ x (t) and h (t).
Compute the discrete convolution sum k=−∞ x (k∆) h ((n − k) ∆)with the built-in LabVIEW MathScript
command conv. Then, multiply this sum by ∆ to get an estimate of y (t) at t = n∆ Note that as ∆ is made
smaller, one gets a closer approximation to y (t).
61
3.1.2 Convolution Properties

Convolution satises the following three properties (see Figure 3.2):
• Commutative property
x (t) ∗ h (t) = h (t) ∗ x (t) (3.7)
• Associative property
x (t) ∗ h1 (t) ∗ h2 (t) = x (t) ∗ {h1 (t) ∗ h2 (t)} (3.8)
• Distributive property
x (t) ∗ {h1 (t) + h2 (t)} = x (t) ∗ h1 (t) + x (t) ∗ h2 (t) (3.9)
Figure 3.2: Convolution Properties

2
3.2 Lab 3: Convolution and Its Applications
This lab involves experimenting with the convolution of two continuous-time signals. The main mathematical
part is written as a .m le, which is then used as a LabVIEW MathScript node within the LabVIEW
programming environment to gain user interactivity. Due to the discrete-time nature of programming, an
approximation of the convolution integral is needed. As an application of the convolution concept, echoes
are removed from speech recordings using this concept.
3.2.1 Numerical Approximation of Convolution

In this section, let us apply the LabVIEW MathScript function conv to compute the convolution of two
signals. One can choose various values of the time interval ∆ to compute numerical approximations to the
convolution integral.
3.2.1.1 Convolution Example 1

In this example, use the function conv to compute the convolution of the signals x (t) = exp (−at) u (t) and
h (t) = exp (−bt) u (t)with u (t)representing a step function starting at 0 for 0 ≤ t ≤ 8. Consider the following
values of the approximation pulse width or delta: ∆ = 0.5, 0.1, 0.05, 0.01, 0.005, 0.001. Mathematically, the
convolution of h (t)and x (t)is given by
1
e−bt − e−at u (t)

y (t) = (3.10)
a−b
Θ
Compare the approximation y (n∆)obtained via the function conv with the theoretical value y (t)given
Θ Θ
by Equation (1). To better see the dierence between the approximated y (n∆)and the true y (n∆)values,
Θ
display y (t)and y (t) in the same graph.
Compute the mean squared error (MSE) between the true and approximated values using the following
equation:
N 2
1 X Θ
MSE = y (n∆) − y (n∆) (3.11)
N n=1
T
where N = b∆ c, T is an adjustable time duration expressed in seconds and the symbol b.c denotes the
nearest integer. To begin with, set T = 8.
As you can see here, the main program is written as a .m le and placed inside LabVIEW as a LabVIEW
MathScript node by invoking Functions → Programming →Structures → MathScript. The .m le
can be typed in or copied and pasted into the LabVIEW MathScript node. The inputs to this program
consist of an approximation pulse width ∆, input exponent powers aand b and a desired time duration T.
To add these inputs, right-click on the border of the LabVIEW MathScript node and click on the Add
Input option as shown in Figure 3.3.
63
Figure 3.3: (a) Adding Inputs, (b) Creating Controls
After adding these inputs, create controls to allow one to alter the inputs interactively via the front
panel. By right-clicking on the border, add the outputs in a similar manner. An important consideration is
the selection of the output data type. Set the outputs to consist of MSE, actual or true convolution output
y_ac and approximated convolution output y. The rst output is a scalar quantity while the other two are
one-dimensional vectors. The output data types should be specied by right-clicking on the outputs and
selecting the Choose Data Type option (see Figure 3.4).
Figure 3.4: (a) Adding Outputs, (b) Choosing Data Types

65
Next write the following .m le textual code inside the LabVIEW MathScript node:
t=0:Delta:8;
Lt=length(t);
x1=exp(-a*t);
x2=exp(-b*t);
y=Delta*conv(x1,x2);
y_ac=1/(a-b)*(exp(-b*t)-exp(-a*t));
MSE=sum((y(1:Lt)-y_ac).^2)/Lt
With this code, a time vector t is generated by taking a time interval of Delta for 8 seconds. Convolve the
two input signals, x1 and x2, using the function conv. Compute the actual output y_ac using Equation (1).
Measure the length of the time vector and input vectors by using the command length(t). The convolution
output vector y has a dierent size (if two input vectors m and n are convolved, the output vector size is
m+n-1). Thus, to keep the size the same, use a portion of the output corresponding to y(1:Lt) during the
error calculation.
Use a waveform graph to show the waveforms. With the function Build Waveform (Functions →
Programming → Waveforms → Build Waveforms), one can show the waveforms across time. Connect
the time interval Delta to the input dt of this function to display the waveforms along the time axis (in
seconds).
Merge together and display the true and approximated outputs in the same graph using the function
Merge Signal (Functions → Express → Sig Manip → Merge Signals). Congure the properties of the
waveform graph as shown in Figure 3.5.
Figure 3.5: Waveform Graph Properties Dialog Box
Figure 3.6 illustrates the completed block diagram of the numerical convolution.
67
Figure 3.6: Block Diagram of the Convolution Example
Figure 3.7 shows the corresponding front panel, which can be used to change parameters. Adjust the
input exponent powers and approximation pulse-width Delta to see the eect on the MSE.
Figure 3.7: Front Panel of the Convolution Example

t−2

Next, consider the convolution of the two signals x (t) = exp (−2t) u (t)and h (t) = rect
2 for , where
u (t)denotes a step function at time 0 and rect a rectangular function dened as
1 −0.5 ≤ t < 0.5

rect (t) ={ (3.12)
0 otherwise
Let ∆ = 0.01. Figure 3.8 shows the block diagram for this second convolution example. Again, the .m le
textual code is placed inside a LabVIEW MathScript node with the appropriate inputs and outputs.
69
Figure 3.8: Block Diagram for the Convolution of Two Signals
Figure 3.9 illustrates the corresponding front panel where x (t), h (t) and x (t)∗h (t) are plotted in dierent
graphs. Convolution (∗) and equal (=)signs are placed between the graphs using the LabVIEW function
Decorations.
Figure 3.9: Front Panel for the Convolution of Two Signals

In this third example, compute the convolution of the signals shown in Figure 3.10.
71
Figure 3.10: Signals x1(t) and x2(t)
Figure 3.11 shows the block diagram for this third convolution example and Figure 3.12 the corresponding
front panel. The signals x1 (t), x2 (t) and x1 (t) ∗ x2 (t) are displayed in dierent graphs.
Figure 3.11: Block Diagram for the Convolution of Two Signals

73
Figure 3.12: Front Panel for the Convolution of Two Signals
3.2.2 Convolution Properties

In this part, examine the properties of convolution. Figure 3.13 shows the block diagram to examine the
properties and Figure 3.14 and Figure 3.15 the corresponding front panel. Both sides of equations are plotted
in this front panel to verify the convolution properties. To display dierent convolution properties within a
limited screen area, use a Tab Control (Controls →Modern→Containers→Tab Control) in the front
panel.
Figure 3.13: Front Panel of Convolution Properties

75
Figure 3.14: Block Diagram of Convolution Properties

Figure 3.15: Tabs Showing Convolution Properties

77
3.2.3 Linear Circuit Analysis Using Convolution

In this part, let us consider an application of convolution in analyzing RLC circuits to gain a better under-
standing of the convolution concept. A linear circuit denotes a linear system, which can be represented with
its impulse response h (t), that is, its response to a unit impulse input. The input to such a system can be
considered to be a voltage v (t)and the output to be the circuit current i (t). See Figure 3.16.
Figure 3.16: Impulse Response Representation of a Linear Circuit
For a simple RC series circuit shown in Figure 3.17, the impulse response is given by [9] ,

1 1
h (t) = exp − t (3.13)
RC RC
which can be obtained for any specied values of R and C. When an input voltage v (t) (either DC or AC)
is applied to the system, the circuit current i (t) can be obtained by simply convolving the system impulse
response with the input voltage, that is
i (t) = h (t) ∗ v (t) (3.14)
Figure 3.17: RC Circuit
Similarly, for the simple RL series circuit shown in Figure 3.18, the impulse response is given by [9] ,

R R
h (t) = exp − t (3.15)
L L
When an input voltage v (t) is applied to the system, the circuit current i (t) can be obtained by computing
the convolution integral.
Figure 3.18: RL Circuit
Figure 3.19 shows the block diagram of this linear system and Figure 3.20 the corresponding front panel.
From the front panel, one can control the system type (RL or RC), input voltage type (DC or AC) and input
voltage amplitude. One can also observe the system response by changing R, L and C values. Three graphs
are used to display the input voltage v (t), impulse response of the circuit h (t) and circuit current i (t).
79
Figure 3.19: Block Diagram of the Linear Circuit Application

Figure 3.20: Front Panel of the Linear Circuit Application

81
3.2.4 Lab Exercises

Echo Cancellation
In this exercise, consider the problem of removing an echo from a recording of a speech signal.
The LabVIEW MathScript function sound() or the function Play Waveform in LabVIEW can
be used to play back the speech recording. To begin, load the .m le echo_1.wav provided on the
book website by using the function wavread(`filename'). This speech le was recorded at the
sampling rate of 8 kHz, which can be played back through the computer speakers by typing
sound(y)
You should be able to hear the sound with an echo. If the LabVIEW function Play Wave-
form(Functions → Programming → Graphics & Sound → Sound→ Output→ Play
Waveform) is used to play the sound, you rst need to build a waveform based on the loaded
data and the time interval dt = 1/8000 because this speech was recorded using an 8 kHz sampling
rate. Connect the waveform to the function Play Waveform.
An echo is produced when the signal (speech, in this case) is reected o a non-absorbing surface
like a wall. What is heard is the original signal superimposed on the signal reected o the wall
(echo). Because the speech is partially absorbed by the wall, it decreases in amplitude. It is also
delayed. The echoed signal can be modeled as ax (t − τ )where a < 1 and τ denotes the echo delay.
Thus, one can represent the speech signal plus the echoed signal as [7]
y (t) = x (t) + ax (t − τ ) (3.16)
What is heard is y (t). In many applications, it is important to recover x (t) the original, echo-free
signal from y (t).
Method 1
In this method, remove the echo using deconvolution. Rewrite Equation (7) as follows [7]:
y [n∆] = x [n∆] + ax [(n − N ) ∆] = x [n∆] ∗ (δ [n∆] + aδ [n − N ] ∆) = x [n∆] ∗ h [n∆] (3.17)
The echoed signal is the convolution of the original signal x (n∆) and the signal h (n∆). Use the
LabVIEW MathScript function deconv(y,h) to recover the original signal.
Method 2
An alternative way of removing the echo is to run the echoed signal through the following
system:
z [n∆] = y [n∆] − az [(n − N ) ∆] (3.18)
Assume that z [n∆] = 0for n < 0. Implement the above system for dierent values of a and N.
Display and play back the echoed signal and the echo-free signal using both of the above methods.
Specify the parameters aand N as controls. Try to measure the proper values of aand N by the
autocorrelation method described below.
The autocorrelation of a signal can be described by the convolution of a signal with its mirror.
That is,
Rxx [n] = x [n] ∗ x [−n] (3.19)
Use the autocorrelation of the output signal (echo-free signal) to estimate the delay time ( N) and
the amplitude of the echo ( a). For dierent values of N and a, observe the autocorrelation output.
To have an echo-free signal, the side lobes of the autocorrelation should be quite low, as shown in
Figure 3.21.
Figure 3.21: Autocorrelation Function of a Signal: (a) Echo Is Not Removed Completely; (b) Echo Is
Removed
Figure 3.22 shows a typical front panel for this exercise. It is not necessary to obtain the same
front panel but there should be controls for a and N as well as graphs to observe the echoed signal,
echo-free signal and autocorrelation function of the echo-free signal.
83
Figure 3.22: Front Panel for the Echo Cancellation System

Noise Reduction Using Mean Filtering
The idea of mean ltering is simply to replace each value in a signal with the mean (average)
value of its neighbors. A mean lter is widely used for noise reduction.
Start by adding some random noise to a signal (use the le echo_1.wav or any other speech data
le). Then, use mean ltering to reduce the introduced noise. More specically, take the following
steps:
1. Normalize the signal values in the range [0 1].

2. Add random noise to the signal by using the function randn. Set the noise level as a control.
3. Convolve the noise-added signal with a mean lter. This lter can be designed by taking an
odd number of ones and dividing by the size. For example, a 1×3 size mean lter is given by
[1/3 1/3 1/3] and a 1 × 5size mean lter by [1/5 1/5 1/5 1/5 1/5]. Set the size of the mean
lter as an odd number control (3, 5 or 7, for example).

Impulse Noise Reduction Using Median Filtering
A median lter is a non-linear lter that replaces a data value with the median of the values
within a neighboring window. For example, the median value for this data stream [2 5 3 11 4] is
4. This type of lter is often used to remove impulse noise. Use the le echo_1.wav or any other
speech data le and take the following steps:
1. Normalize the signal values in the range [0 1].

2. Randomly add impulse noise to the signal by using the LabVIEW MathScript function
randperm. Set the noise density as a control.
3. Find the median values of neighboring data using the function median and replace the original
value with the median value. Set the number of neighboring values as an odd number control
(3, 5 or 7, for example).
85
Solutions to Lab 3: Convolution and Its Applications

Chapter 4
Fourier Series
1
4.1 Fourier Series
4.1.1 Background
A periodic signal x (t)can be expressed by an exponential Fourier series as follows:
∞
2π nt
X
x (t) = cn ej T (4.1)
n=−∞
where T indicates the period of the signal and cn 's are called Fourier series coecients, which, in general,
are complex. Obtain these coecients by performing the following integration
Z
1 2π nt
cn = x (t) e−j T dt (4.2)
T
T
which possesses the following symmetry properties
| c−n |=| cn | (4.3)
∠c−n = −∠cn (4.4)
where the symbol | . | denotes magnitude and ∠ phase. Magnitudes of the coecients possess even symmetry
and their phases odd symmetry.
A periodic signal x (t) can also be represented by a trigonometric Fourier series as follows:
∞
X 2π nt 2π nt
x (t) = a0 + an cos + bn sin (4.5)
n=1
T T
where
Z
1
a0 = x (t) dt (4.6)
T
T
Z
2 2π nt
an = x (t) cos dt (4.7)
T T
T
87
88 CHAPTER 4. FOURIER SERIES
Z
2 2π nt
bn = x (t) sin dt (4.8)
T T
T
The relationships between the trigonometric series and the exponential series coecients are given by
a0 = c0 (4.9)
an = 2Re{cn } (4.10)
bn = −2Im{cn } (4.11)
1
cn = (an − jbn ) (4.12)
2
where Re and Imdenote the real and imaginary parts, respectively.
According to the Parseval's theorem, the average power in the signal x (t) is related to the Fourier series
coecients cn 's, as indicated below
Z ∞
1 2
X 2
| x (t) | dt = | cn | (4.13)
T n=−∞
T
More theoretical details of Fourier series are available in signals and systems textbooks [9] - [15] .
4.1.2 Fourier Series Numerical Computation

Fourier series coecients are often computed numerically in particular, when an analytic expression for
x (t) is not available or the integration in (4.6) - (4.8) is dicult to perform. By approximating the integrals
in (4.6) - (4.8) with a summation of rectangular strips, each of width ∆t, one can write
M
1 X
a0 = x (m∆t) (4.14)
M m=1
M
2 X 2π mn
an = x (m∆t) cos (4.15)
M m=1 M
M
2 X 2π mn
bn = x (m∆t) sin (4.16)
M m=1 M
where x (m∆t) are M x (t) over a single period T , and ∆t denotes

equally spaced data points representing
T
the interval between data points such that
M ∆t =
Similarly, by approximating the integrals in (4.2) with a summation of rectangular strips, each of width
∆t, one can write
M
1 X j2π mn
cn = x (m∆t) exp (4.17)
M M
m=M
89
2
4.2 Lab 4: Fourier Series and Its Applications
In this lab, we examine the representation of periodic signals based on Fourier series. Periodic signals can
be represented by a linear combination of an innite sum of sine waves, as expressed by the trigonometric
Fourier series representation.Periodic signals can also be represented by an innite sum of harmonically
related complex exponentials, as expressed by the exponential Fourier series representation. In this lab,
we analyze both of these series representations. In particular, we focus on how to compute Fourier series
coecients numerically.
4.2.1 Fourier Series Signal Decomposition and Reconstruction

This example helps one to gain an understanding of Fourier series decomposition and reconstruction for
periodic signals. The rst step involves estimating x (m∆t) which is a numerical approximation of the
analog input signal. Though programming environments deploy discrete values internally, we can obtain a
close analog approximation of a continuous-time signal by using a very small ∆t. That is to say, for all
practical purposes, when ∆t is taken to be very small, we get the analog representation or simulation of the
signal. In this example, create four input signals using the listed LabVIEW MathScript functions in Table
1.
Waveform type LabVIEW MathScript function

Square wave square(T), T denotes period
Triangular wave sawtooth(T,Width), Width=0.5
Sawtooth wave sawtooth(T,Width), Width=0
sin (2 ∗ pi ∗ f ∗ t) for 0 ≤ t < T /2

Half wave rectied sine wave { , f =
0 for T /2 ≤ t < T
1/T denotes frequencyHalf period is sine wave and
the other half is made zero
Table 4.1: LabVIEW MathScript Functions for Generating Various Waveforms or Signals
Use a switch structure to select dierent types of input waveforms. Set the switch parameter w as the
input and connect it to an Enum Control(Controls → Modern → Ring & Enum → Enum). Edit
the Enum Control items to include all the waveform types.
Set Amplitude of input (A), Period of input (T) and Number of Fourier coecients (N) as control pa-
rameters. Determine Fourier coecients a0 , an and bn and reconstruct the signal from its Fourier coecients
using equations provided in Chapter 4. Determine the error between the input and the reconstructed signal
Θ
by simply taking the absolute values of x (t) − x (t)via the LabVIEW MathScript function abs. Finally,
determine the maximum and average errors by using the functions max and sum. Figure 4.1 shows the
completed block diagram of the Fourier series signal decomposition and reconstruction system.
Display the input signal using a waveform graph. Before displaying the graph, congure it using the
function Build Waveform(Functions → Programming → Waveforms → Build Waveforms). Also
display the Fourier coecients, reconstructed signal and error in the waveform graph, and place several
numerical indicators to show the values of the Fourier coecients, maximum error and average error.
Figure 4.2 and Figure 4.3 illustrate the front panel of the Fourier series signal decomposition and re-
construction system, respectively. To display all the outputs within a limited screen area, use a Tab Con-
trol(Controls → Modern → Containers → Tab Control) in the front panel. Here the outputs are
arranged in two dierent tabs: Fourier Series and Signal Reconstruction.

Figure 4.1: Block Diagram of Fourier Series Signal Decomposition and Reconstruction Example
91
Figure 4.2: Front Panel of Fourier Series Signal Decomposition and Reconstruction Example (Fourier
Series Tab)
93
Figure 4.3: Front Panel of Fourier Series Signal Decomposition and Reconstruction Example (Signal
Reconstruction Tab)
95
4.2.2 Linear Circuit Analysis Using Trigonometric Fourier Series

In this example, let us perform electrical circuit analysis using the trigonometric Fourier series. The ability
to decompose any periodic signal into a number of sine waves makes the Fourier series a powerful tool in
electrical circuit analysis. The response of a circuit component when a sinusoidal input is applied to its
terminals is well-known in circuit analysis. Thus, to obtain the response to any periodic signal, one can
decompose the signal into sine waves and perform a linear superposition of the sine waves.
Consider a simple RC circuit excited by a periodic input signal as shown in Figure 4.4.
Figure 4.4: RC Series Circuit with Periodic Input Voltage
The block diagram of this linear circuit or system is shown in Figure 4.5. Determine the Fourier series
coecients of the input voltage signal as discussed in the previous example. Because the Fourier series
involves the sum of sinusoids, phasor analysis can be used to obtain the output voltage (vc ). Let n represent
the number of terms in the Fourier series. By using the voltage divider rule, the output voltage (vc ) can be
expressed as [9],
1/ (jnωC)
vc n = vin (4.18)
R + 1/ (jnωC) n
Because the sine and cosine components of the input voltage are known, one can easily determine the output
by adding the individual output components because the circuit is linear. Determine each output voltage
component by using (4.18).
Figure 4.6 and Figure 4.7 show the front panel of this system for its two tabs. The magnitude and phase
of the sine and cosine components are shown in the front panel separately. Furthermore, the tab control is
used to show the Fourier series and system output separately.
Figure 4.5: Block Diagram of Circuit Analysis with Trigonometric Fourier Series
97
Figure 4.6: Front Panel of Circuit Analysis with Trigonometric Fourier Series (Fourier Series Tab)
99
Figure 4.7: Front Panel of Circuit Analysis with Trigonometric Fourier Series (System Output Tab)
101
4.2.3 Lab Exercises

RL Circuit Analysis
Build a hybrid VI to analyze the RL circuit shown in Figure 4.8 using Fourier series.
Figure 4.8: RL Series Circuit with Periodic Input Voltage
The input voltage for the circuit is to be either a square wave or a triangular wave with a period
T=2 seconds.
Compute and display the following:
1. The Fourier series coecients of the input voltage v (t),

2. the current i (t),
3. the RMS (root mean square) value of v (t) using (i) the original waveform and (ii) its Fourier
series coecients (compare the outcomes),
4. the average power Pav delivered by the source.
Hints:
RMS Value
The RMS value of a periodic function v (t)with period T is given by
v Z
u1
VRMS = t v 2 dt (4.19)
u
T
T
The RMS value of a waveform consisting of sinusoids with dierent frequencies is equal to the square
root of the sum of the squares of the RMS value of each sinusoid. If a waveform is represented by
the following Fourier series
v (t) = V0 + V1 sin (ω1 t ± φ1 ) + V2 sin (ω2 t ± φ2 ) + ... + VN sin (ωN t ± φN ) (4.20)
then, the RMS value VRMS is given by
s 2 2 2
V V VN
VRMS = V02 + √1 + √2 + ... + √ (4.21)
2 2 2
Average power
The average power of the Fourier series can be expressed as
Pav = V0 I0 + V1RMS I1RMS cosφ1 + V2RMS I2RMS cosφ2 + ... (4.22)


Doppler Eect
The Doppler eect denotes the change in frequency and wavelength of a wave as perceived by
an observer moving relative to the wave source. The Doppler eect can be demonstrated via time
scaling of Fourier series. The observer hears the siren of an approaching emergency vehicle with
dierent amplitudes and frequencies as compared to the original signal. As the vehicle passes by, the
observer hears another amplitude and frequency. The reason for the amplitude change (increased
loudness) is because of the proximity of the vehicle. The closer it is, the louder it gets. The reason
for frequency (pitch) change is due to the Doppler eect. As the vehicle approaches, each successive
compression of the air caused by the siren occurs a little closer than the last one, and the opposite
happens when the vehicle passes by. The result is the scaling of the original signal in the time
domain, which changes its frequency. When the vehicle approaches, the scaling factor is greater
than 1, resulting in a higher frequency, and, when it passes by, the scaling factor is less than 1,
resulting in a lower frequency. More theoretical aspects of this phenomenon are covered in reference
[16].
Dene the original siren signal as x (t). When the vehicle approaches, one can describe the
signal by
x1 (t) = B1 (t) x (at) (4.23)
where B1 (t)is an increasing function of time (assuming a linear increment with time) and ais the
scaling factor having a value greater than 1. When the vehicle passes by, one can describe the signal
by
x2 (t) = B2 (t) x (bt) (4.24)
where B2 (t)is a decreasing function of time (assuming a linear decrement with time) and bis the
scaling factor having a value less than 1.
First, generate a signal and create an upscale and a downscale version of it. Observe the Fourier
series for all the signals. Set the amplitude and frequency of the original signal and the scaling
factors as controls. In addition, play the sounds using the LabVIEW Play Waveform function.
Figure 4.9 shows a possible front panel for this type of system.
103
Figure 4.9: Front Panel of a Doppler Eect System

Synthesis of Electronic Music
In electronic music instruments, sound generation is implemented via synthesis. Dierent types
of synthesis techniques such as additive synthesis, subtractive synthesis and frequency modulation
(FM) synthesis are used to create audio waveforms. The simplest type of synthesis is additive
synthesis, where a composite waveform is created by summing sine wave components, which is
basically the inverse Fourier series operation. However, in practice, to create a music sound with
rich harmonics requires adding a large number of sine waves, which makes the approach inecient
computationally. To avoid adding a large number of sine waves, modulation with addition is used.
This exercise involves the design of algorithms used in the Yamaha DX7 music synthesizer, which
debuted in 1983 as the rst commercially available digital synthesizer.
The primary functional circuit in DX7 consists of a digital sine wave oscillator plus a digital
envelope generator. Let us use additive synthesis and frequency modulation to achieve synthesis
with six congurable operators. When one adds together the output of some operators, an additive
synthesis occurs, and when one connects the output of one operator to the input of another operator,
a modulation occurs.
In terms of block diagrams, the additive synthesis of a waveform with four operators is illustrated
in Figure 4.10.
Figure 4.10: Additive Synthesis
The output for the combination shown in Figure 4.10 can be written as
y (t) = A1 sin (ω1 t) + A2 sin (ω2 t) + A3 sin (ω3 t) + A4 sin (ω4 t) (4.25)
Figure 4.11 shows the FM synthesis of a waveform with two operators.
Figure 4.11: FM Synthesis
The output for the combination shown in this gure can be written as
y (t) = A1 sin (ω1 t + A2 sin (ω2 t)) (4.26)
Other than addition and frequency modulation, one can use feedback or self-modulation in DX7,
which involves wrapping back and using the output of an operator to modulate the input of the
same operator as shown in Figure 4.12.
105
Figure 4.12: Self-Modulation
The corresponding equation is
y (t) = A1 sin (ω1 t + y (t)) (4.27)
Dierent arrangements of operators create dierent algorithms. Figure 4.13 displays the diagram
of an algorithm.
Figure 4.13: Diagram of an Algorithm
And the output for this algorithm can be written as
y (t) = A1 sin (ω1 t + A2 sin (ω2 t)) + A3 sin (ω3 t + A4 sin (ω4 t + A5 sin (ω5 t + y6 (t)))) (4.28)
With DX7, one can choose from 32 dierent algorithms. As one moves from algorithm No. 32
to algorithm No. 1, the harmonics complexity increases. In algorithm No. 32, all six operators
are combined using additive synthesis with a self modulator generating the smallest number of
harmonics. Figure 4.14 shows the diagram for all 32 combinations of operators. More details on
music synthesis and the Yamaha DX7 synthesizer can be found in the [1]-[20].
Next, explore designing a system with six operators and set their amplitude and frequency as
controls. By combining these operators, construct any three algorithms, one from the lower side
(for example, algorithm No. 3), one from the middle side (for example, algorithm No. 17) and the
nal one from the upper side (for example, algorithm No. 30). Observe the output waves in the
time and frequency domains (nd the corresponding Fourier series).
Figure 4.14: 32 Algorithms in the Yamaha DX7

107
Solutions to Lab 4: Fourier Series and Its Applications

Chapter 5
Continuous-Time Fourier Transform
1
5.1 Continuous-Time Fourier Transform
In this lab, we learn how to compute the continuous-time Fourier transform (CTFT), normally referred to as
Fourier transform, numerically and examine its properties. Also, we explore noise cancellation and amplitude
modulation as applications of Fourier transform.
5.1.1 Properties of CTFT

The continuous-time Fourier transform (CTFT) (commonly known as Fourier transform) of an aperiodic
signal x (t) is given by
Z∞
X (ω) = x (t) e−jωt dt (5.1)
−∞
The signal x (t) can be recovered from X (ω) via this inverse transform equation
Z∞
1
x (t) = X (ω) ejωt dω (5.2)
2π
−∞
Some of the properties of CTFT are listed in Table 5.1.
Properties Time domain Frequency domain

Time shift x (t − t0 ) X (ω) e−jωt0
1 ω

Time scaling x (at) |a| X a
Linearity a1 x1 (t) + a2 x2 (t) a1 X1 (ω) + a2 X2 (ω)
Time convolution x (t) ∗ h (t) X (ω) H (ω)
Frequency convolution x (t) h (t) X (ω) ∗ H (ω)
Table 5.1: Properties of CTFT
Refer to signals and systems textbooks [9] - [15] for more theoretical details on this transform.
109
110 CHAPTER 5. CONTINUOUS-TIME FOURIER TRANSFORM
5.1.2 Numerical Approximations to CTFT

Assuming that the signal x (t) is zero for t<0 and t ≥ T, we can approximate the CTFT integration in
Equation (1) as follows:
Z∞ ZT N
X −1
−jωt
x (t) e dt = x (t) e−jωt dt ≈ x (nτ ) e−jωnτ τ (5.3)
−∞ 0 n=0
where T = Nτ and N is an integer. For suciently small τ , the above summation provides a close approxi-
PN −1
mation to the CTFT integral. The summation n=0 x (nτ ) e−jωnτ is widely used in digital signal processing
(DSP), and both LabVIEW MathScript and LabVIEW have a built-in function for it called fft. In a .m
le, if N samples x (nτ ) are stored in a vector x, then the function call
xw=tau*fft (x)
calculates
PN −1
Xω (k + 1) = τ n=0 x (nτ ) e−jωk nτ 0≤k ≤N −1
(5.4)
≈ X (ωk )
where
2πk N
Nτ 0≤k≤ 2
ωk = { (5.5)
2πk 2π N
Nτ − τ 2 +1≤k ≤N −1
with N assumed to be even. The fft function returns the positive frequency samples before the negative
frequency samples. To place the frequency samples in the right order, use the function fftshift as indicated
below:
xw=fftshift(tau*fft (x ) )
Note that X (ω) is a vector (actually, a complex vector) of dimension N. X (ω) is complex in general
despite the fact that x (t) is real-valued. The magnitude of X (ω) can be computed using the function abs
and the phase of X (ω) using the function angle.
111
2
5.2 Lab 5: CTFT and Its Applications
5.2.1 Properties of CTFT

The example covered in this section provides an implementation of CTFT and its properties. As mentioned
earlier, programming environments can generate and work with only discrete values arranged in arrays.
Thus, to get a continuous-time representation of a signal, use a very small value of time increment dt. For
example, dt=0.001 means there are 1000 discrete samples in 1 second, which provides a good approximation
to represent a low-frequency signal. However, when working with very high-frequency signals, one should
decrease the value of dt further to ensure there are enough samples to represent the signal in a continuous
fashion over a specied duration.
Figure 5.1 shows the example of the completed block diagram for the CTFT (or FT) and its properties.
This particular VI is capable of nding the FT of a rectangular and a triangular pulse. Create two input
signals using the LabVIEW MathScript functions ones and zeros, which are combined in the time domain.
Use a case structure to select the combination method (linear combination, convolution or multiplication)
and the parameter mode1 to serve as an input that is connected to an Enum Control(Controls →
Modern→Ring & Enum →Enum). Use parameters mode3 and mode4, which are connected to two
Enum controls, to select the input signal type. Also set Pulse width, Time shift and Time scale as control
parameters. Pulse width controls the number of ones in the pulse, which is used to increase or decrease
the pulse width. Time shift adds zeros before the pulse to provide a time delay. Time scale is set to be
multiplied with the time increment (dt) to ensure appropriate scaling of the pulse. Use the LabVIEW
MathScript function fft to determine the FT of the continuous signal. Combine the signals in the frequency
domain and control the combination method (linear combination, convolution or multiplication) via the
parameter mode2. Compute the FT of the time domain combinations and the inverse FT of the frequency
domain combinations using the functions fft and ifft. To shift the zero-frequency component to the
center of the spectrum, use the LabVIEW MathScript function fftshift. Finally, determine the magnitude
and phase of the FT using the functions abs and angle, respectively. Display the input signals and their
combinations using a Build Waveform function (Functions → Programming →Waveforms →Build
Waveforms) and a Waveform Graph(Controls →Modern →Graph →Waveform Graph). Also,
display the spectrum magnitude and phase using a waveform graph.

Figure 5.1: Block Diagram of CTFT and Its Properties

113
Figure 5.2 and Figure 5.3 shows the front panel of the above system. Use controls named Pulse width,
Time shift and Time scaling to change the waveforms in the time domain. Three waveform graphs for Input
signal, Magnitude of FT and Phase of FT also appear in the front panel shown. With the specied front
panel controls, one can easily verify CTFT properties. To begin with, run the program in continuous mode
using the Run Continuously button.
Figure 5.2: Front Panel of CTFT and Its Properties: Input Signals Tab
115
Figure 5.3: Front Panel of CTFT and Its Properties: Combination of Input Signals Tab
117
5.2.1.1 Varying Pulse Width

Keep the default values of Time shift (=0) and Time scaling (=1) and vary the Pulse width of the rectangular
pulse. First, set the value of the Pulse width to its minimum value (=0.01) and then increase it. Observe
that increasing the Pulse width in the time domain decrements the width in the frequency domain (see
Figure 5.4 ). When the Pulse width is set to its maximum value (=1) in the frequency domain, only one
value can be seen at the center frequency indicating the signal is of DC type (refer to Properties of CTFT
section of Chapter 5).
Figure 5.4: Magnitude Spectrum for Dierent Pulse Widths: (a) 0.01, (b) 0.2, (c) 0.5, (d) 1
119
5.2.1.2 Time Shift

Next, for a xed pulse width, vary the time shift. Observe that the phase spectrum changes but the
magnitude spectrum remains the same. If the signal x (t) is shifted by a constant t0 , its FT magnitude does
not change, but the term −ωt0 gets added to its phase angle. This veries the time-shifting property of FT
as stated in Properties of CTFT section of Chapter 5 (see Figure 5.5).
Figure 5.5: Magnitude and Phase Spectrum for Dierent Time Shifts: (a) 0, (b) 0.2, (c) 0.5, (d) 0.7
121
5.2.1.3 Time Scaling

Observe that increasing the control Time scaling makes the spectrum wider. This indicates that compressing
the signal in the time domain leads to expansion in the frequency domain. This veries the time-scaling
property of FT as stated in Properties of CTFT section of Chapter 5 (see Figure 5.6).
Figure 5.6: Magnitude Spectrum for Dierent Time Scalings: (a) 1, (b) 2, (c) 3, (d) 4
123
5.2.1.4 Linearity
Here, combine two signals to examine the linearity property of FT. Select Linear Combination for the Time
domain and Frequency domain combination method. This selection combines two time signals, x1 (t)and
x2 (t), linearly with the scaling factors, a1 and a2 , producing a new signal, a1 x1 (t) + a2 x2 (t). Figure 5.7
displays the FT of this linear combination. The linear combination in the frequency domain produces a
new signal, a1 X1 (ω) + a2 X2 (ω). Figure 5.7 also displays the inverse FT of this combination. Observe that
both combinations produce the same result in the time and frequency domains, as indicated by the linearity
property stated in Properties of CTFT section of Chapter 5.
Figure 5.7: Verifying the Linearity Property of CTFT

125
5.2.1.5 Time Convolution

In this part, convolve two signals in the time domain to examine the time-convolution property of FT.
Select Convolution for Time domain and Multiplication for Frequency domain. This selection produces and
displays a new signal, x1 (t) ∗ x2 (t), by convolving the two time signals x1 (t)and x2 (t). Multiplication in
the frequency domain produces a new signal, X1 (ω) X2 (ω). The inverse FT of this multiplied signal is also
displayed on the right. Note that both combinations produce the same outcome in the time and frequency
domains. This veries the time-convolution property stated in the Properties of CTFT section of Chapter 5
(see Figure 5.8).
Figure 5.8: Verifying the Time-Convolution Property of CTFT

127
5.2.1.6 Frequency Convolution

Convolve two signals in the frequency domain to examine the frequency-convolution property of FT. Select
Convolution for Frequency domain and Multiplication for Time domain. This selection convolves the two
time signals X1 (ω)and X2 (ω) to produce a new signal, X1 (ω) ∗ X2 (ω). The inverse FT of the convolved
signal is displayed. Multiplication in Time domain produces a new signal, x1 (t) x2 (t). The FT of this
multiplied signal is also displayed. Note that both combinations produce the same outcome in the time
and frequency domains. This veries the frequency-convolution property stated in the Properties of CTFT
section of Chapter 5 (see Figure 5.9).
Figure 5.9: Verifying the Frequency-Convolution Property of CTFT

129
5.2.2 Noise Reduction

When a signal passes through a channel, it normally gets corrupted by channel noise. Various electronic
components used in a transmitter or receiver may also cause additional noise. Noise reduction is an important
aspect of any signal processing system. Lab 7 features noise reduction techniques using digital and analog
lters. This section presents a simple technique to reduce high-frequency noise.
Figure 5.10 shows the completed block diagram of a noise reduction system. Consider a speech signal
sampled at 8 kHz. Add some high-frequency noise to this signal and then remove the high-frequency compo-
nents in the frequency domain. Finally, move the signal back into the time domain using the inverse FT. Use
the LabVIEW MathScript function wavread to read a wave le specied by the string Path and return the
sampled data at a specied sampling rate. A String Control(Controls → Modern →String & Path
→String Control) can be added to the input Path to provide the path name for the speech data le. Use
two more controls named Time frame width and Frame number to extract a segment of the speech signal
before computing Fourier transform. Add together three sine and cosine waves with frequencies of 3.5, 3 and
2.8 kHz to create a high-frequency noise. Then add a scaled version of the noise signal to the signal with
the Scaling parameter set as a control. Compute the FT of the Noise added signal using the function fft.
To remove the high-frequency noise components, use a simple lowpass lter by removing the frequency
components over a certain threshold (50 percent, for example). After removing the high-frequency com-
ponents, transform the signal back into the time domain using the function ifft. To get a display of the
absolute and centered frequency spectrum, use the functions abs and fftshift. The signals are displayed
Build Waveform and Waveform Graph. To be able to hear the
in the time domain using the functions
Play Waveform(Functions →Programming →Graphics & Sound
speech signals, use the function
→Sound→Output →Play Waveform). Connect the time domain signals to this function via the while
loop structure. Connect a Boolean control(Controls →Modern →Boolean →Push Button) to the
loop control, which acts as a play switch for the sound signal. The signals are also displayed in the fre-
Bundle(Functions →Programming →Cluster, Class & Variant
quency domain using the functions
→ Bundle) and XY Graph(Controls →Modern →Graphs →XY Graph).
Figure 5.10: Block Diagram of a Noise Reduction System

131
Figure 5.11 shows the front panel of the system. Inside the Data File Path control, the location of the
speech data le is specied. Three graphs for the Original signal, Noise added signal and Noise reduced
signal are shown in both the time and frequency domains. Use the noise level control to allow setting the
amount of noise added to the original signal. After running the program, click on the Play button next to
each signal. Hear the Original Signal and the Noise added signal. Notice that an unpleasant high-pitched
noise gets added to the signal. If the noise level is set more than 0.5, the Original Signal becomes very
dicult to hear. Next, hear the Noise reduced signal, which is similar to the Original Signal. The Noise
reduced signal is not exactly the same as the Original Signal because some high-frequency components are
also removed along with the noise.
Figure 5.11: Front Panel of a Noise Reduction System

133
5.2.3 Amplitude Modulation

In this section, we examine amplitude modulation and demodulation applications. For transmission purposes,
signals are often modulated with a high-frequency carrier. A typical amplitude modulated signal can be
described by
x (t) = xm (t) cos (2πf c t) (5.6)
where xm (t)is called the message waveform, which contains the data of interest, and fc is the carrier wave
frequency. Using the fact that
1 2πf c t 1 ωc t
+ e−2πf c t = e + e−ωc t

cos (2πf c t) = e (5.7)
2 2
and the frequency shift property of CTFT, one can easily derive the CTFT of to be
1
X (ω) = (Xm (ω − ωc ) + Xm (ω + ωc )) (5.8)
2
At the receiver, some noisy version of this transmitted signal is received. The signal information resides
in the envelope of the modulated signal, and thus an envelope detector can be used to recover the message
signal.
Figure 5.12 shows the completed block diagram of the amplitude modulation and demodulation system.
In this example, use the combination of two sine waves to serve as a message signal. The signal is modulated
with a high-frequency carrier, and some random noise is added. The frequency domain versions of the signals
can also be observed using the function fft. As stated in Equation (3), the CTFT of the modulated signal
is merely some frequency-shifted version of the original signal. In single sideband (SSB) modulation, only
one side of the spectrum is transmitted due to symmetry. That is, just one side of the spectrum is taken
and converted into a time signal using the function ifft.
Figure 5.13 shows the completed front panel of this system. The Message signal, Modulated signal,
Received signal (modulated signal with additional noise) and Demodulated signal are displayed in four
waveform graphs in both the time and frequency domains.
Figure 5.12: Block Diagram of an Amplitude Modulation and Demodulation System

135
Figure 5.13: Front Panel of an Amplitude Modulation and Demodulation System

5.2.4 Lab Exercises

Circuit Analysis
Find and plot the frequency response (both magnitude and phase spectrum) of each of the
circuits shown in Figure 5.14. Set the values of R, L and C as controls.
Figure 5.14: Linear RLC Circuits

Morse Coding
Consider a message containing some hidden information. Furthermore, to make it interesting,
suppose the message contains a name. Assume that the message was coded using the amplitude
137
modulation scheme as follows [17]:
x (t) = xm1 (t) cos (2πf 1 t) + xm2 (t) cos (2πf 2 t) + xm3 (t) cos (2πf 3 t) (5.9)
where xm1 (t) , xm2 (t) and xm3 (t) are the (message) signals containing the three letters of the
name. More specically, each of the signals, xm1 (t) , xm2 (t) and xm3 (t), corresponds to a single
letter of the alphabet. These letters are encoded using the International Morse Code as indicated
below [7]:
A .− H .... O −−− V ...−
B −... I .. P . − −. W . −−
C −. − . J . − −− Q − − .− X −..−
D −.. K −.− R . −. Y −. − −
E . L . − .. S ... Z − − ..
F .. −. M −− T −
G −−. N −. U ..−
Now to encode the letter A, one needs only a dot followed by a dash. That is, only two prototype
signals are needed one to represent the dash and one to represent the dot. Thus, for instance,
to represent the letter A, set xm1 (t) = d (t) + dash (t), where d (t) represents the dot signal and
dash (t) the dash signal. Similarly, to represent the letter O, set xm1 (t) = 3dash (t).
Find the prototype signals d (t) and dash (t) in the le morse.mat on the book website. After
loading the le morse.mat
load morse
the signals d (t) and dash (t)can be located in the vectors dot and dash, respectively. The hidden
signal, which is encoded, per Equation (4), containing the letters of the name, is in the vector xt
Let the three modulation frequencies f1 , f2 and f3 be 20, 40 and 80 Hz, respectively.
• Using the amplitude modulation property of the CTFT, determine the three possible letters
and the hidden name. (Hint: Plot the CTFT of xt Use the values of T and τ au contained in the
le.)
• Explain the strategy used to decode the message. Is the coding technique ambiguous? That is,
is there a one-to-one mapping between the message waveforms ( xm1 (t) , xm2 (t) , xm3 (t)) and the
alphabet letters? Or can you nd multiple letters that correspond to the same message waveform?

Doppler Eect
The Doppler eect phenomenon was covered in a previous chapter. In this exercise, let us
examine the Doppler eect with a real sound wave rather than a periodic signal. The wave le
retrucksiren.wav on the book website contains a retruck siren. Read the le using the LabVIEW
MathScript function wavread and produce its upscale and downscale versions. Show the waves
in the time and frequency domains (nd the CTFT). Furthermore, play the sounds using the
LabVIEW function Play Waveform. Figure 5.15 shows a typical front panel for this system.
Figure 5.15: Front Panel of Doppler Eect System

Diraction of Light
The diraction of light can be described as a Fourier transform [15]. Consider an opaque
screen with a small slit being illuminated by a normally incident uniform light wave, as shown in
Figure 5.16.
139
Figure 5.16: Diraction of Light
Considering that dπl12 /λprovides a good approximation for any l1 in the slit, the electric eld
strength of the light striking the viewing screen can be expressed as [3]
Z∞
ej(2πd/λ) j(π/λd)l02
E0 (l0 ) = K e E1 (l1 ) e−j(2π/λd)l0 l1 dl1 (5.10)
jλd
−∞
where
E1 = eld strength at diraction screen
E0 = eld strength at viewing screen
K = constant of proportionality
λ= wavelength of light
The above integral is in fact Fourier transformation in a dierent notation. One can write the
eld strength at the viewing screen as [3]
(t)
E1f →l0 /λd
(5.11)
j(2πd/λ)
E0 (l0 ) = K e jλd ej(π/λd)l02 CTFT
The intensity I (l0 )of the light at the viewing screen is the square of the magnitude of the eld
strength. That is,
2
I (l0 ) = | E0 (l0 ) | (5.12)
1. Plot the intensity of the light at the viewing screen. Set the slit width to this range (0.5 to 5
mm), the wavelength of light λto this range (300 to 800 nm), and the distance of the viewing
screen dto this range (10 to 200 m) as controls. Assume the constant of proportionality is
−3
10 , and the electric eld strength at the diraction screen is 1 V/m.
2. Now replace the slit with two slits, each 0.1 mm in width, separated by 1 mm (center-to-
center) and centered on the optical axis. Plot the intensity of light in the viewing screen by
setting the parameters in part (1) as controls.
Solutions to Lab 5: CTFT and Its Applications

Chapter 6
Digital Signals and Their Transforms
1
6.1 Digital Signals and Their Transforms
In this lab, we learn how to compute the continuous-time Fourier transform (CTFT), normally referred to as
Fourier transform, numerically and examine its properties. Also, we explore noise cancellation and amplitude
modulation as applications of Fourier transform.
In the previous labs, dierent mathematical transforms for processing analog or continuous-time signals
were covered. Now let us explore the mathematical transforms for processing digital signals. Digital signals
are sampled (discrete-time) and quantized version of analog signals. The conversion of analog-to-digital
signals is implemented with an analog-to-digital (A/D) converter, and the conversion of digital-to-analog
signals is implemented with a digital-to-analog (D/A) converter. In the rst part of the lab, we learn how
to choose an appropriate sampling frequency to achieve a proper analog-to-digital conversion. In the second
part of the lab, we examine the A/D and D/A processes.
6.1.1 Sampling and Aliasing

Sampling is the process of generating discrete-time samples from an analog signal. First, it is helpful
to mention the relationship between analog and digital frequencies. Consider an analog sinusoidal signal
x (t) = Acos (ωt + φ). Sampling this signal at t = nTs , with the sampling time interval of Ts , generates the
discrete-time signal
x [n] = Acos (ω nTs + φ) = Acos (θn + φ) , n = 0, 1, 2, ..., (6.1)
2πf
where θ = ωTs =
fs denotes digital frequency with units being radians (as compared to analog frequency
ω with units being radians/second).
The dierence between analog and digital frequencies is more evident by observing that the same discrete-
time signal is obtained from dierent continuous-time signals if the product ωTs remains the same. (An
example is shown in Figure 6.1.) Likewise, dierent discrete-time signals are obtained from the same analog
or continuous-time signal when the sampling frequency is changed. (An example is shown in Figure 6.2.)
In other words, both the frequency of an analog signal f and the sampling frequency fs dene the digital
frequency θof the corresponding digital signal.
141
142 CHAPTER 6. DIGITAL SIGNALS AND THEIR TRANSFORMS
Figure 6.1: Sampling of Two Dierent Analog Signals Leading to the Same Digital Signal
Figure 6.2: Sampling of the Same Analog Signal Leading to Two Dierent Digital Signals
It helps to understand the constraints associated with the above sampling process by examining signals
in the frequency domain. The Fourier transform pairs for analog and digital signals are stated as
143
R∞
X (jω) = −∞
x (t) e−jωt dt
Fourier transform pair for analog signals { R∞
1
x (t) = 2π −∞
X (jω) ejωt dω
X ejθ = n=−∞ x [n] e−jnθ , θ = ωTs
P ∞
Fourier transform pair for discrete signals {
X ejθ ejnθ dθ
1
Rπ
x [n] = 2π −π
Table 6.1: Fourier transform pairs for analog and digital signals
Figure 6.3: (a) Fourier Transform of a Continuous-Time Signal, (b) Its Discrete-Time Version
As illustrated in Figure 6.3, when an analog signal with a maximum bandwidth of W (or a maximum
frequency of fmax ) is sampled at a rate of Ts = f1s , its corresponding frequency response is repeated every
2π radians, or fs . In other words, the Fourier transform in the digital domain becomes a periodic version of
the Fourier transform in the analog domain. That is why, for discrete signals, one is interested only in the
frequency range [0, fs /2].
Therefore, to avoid any aliasing or distortion of the discrete signal frequency content and to be able to
recover or reconstruct the frequency content of the original analog signal, we must have fs ≥ 2f max . This
is known as the Nyquist rate. The sampling frequency should be at least twice the highest frequency in the
analog signal. Normally, before any digital manipulation, a front-end anti-aliasing lowpass analog lter is
used to limit the highest frequency of the analog signal.
Let us further examine the aliasing problem by considering an undersampled sinusoid as depicted in
Figure 6.4 . In this gure, a 1 kHz sinusoid is sampled at fs = 0.8kHz, which is less than the Nyquist rate
of 2 kHz. The dashed-line signal is a 200 Hz sinusoid passing through the same sample points. Thus, at the
sampling frequency of 0.8 kHz, the output of an A/D converter is the same if one uses the 1 kHz or 200 Hz
sinusoid as the input signal. On the other hand, oversampling a signal provides a richer description than
that of the signal sampled at the Nyquist rate.
Figure 6.4: Ambiguity Caused by Aliasing
6.1.2 Quantization
An A/D converter has a nite number of bits (or resolution). As a result, continuous amplitude values get
represented or approximated by discrete amplitude levels. The process of converting continuous into discrete
amplitude levels is called quantization. This approximation leads to errors called quantization noise. The
input/output characteristic of a 3-bit A/D converter is shown in Figure 6.5 to illustrate how analog voltage
values are approximated by discrete voltage levels.
145
Figure 6.5: Characteristic of a 3-Bit A/D Converter: (a) Input/Output Transfer Function, (b) Additive
Quantization Noise
Quantization interval depends on the number of quantization or resolution levels, as illustrated in Fig-
ure 6.6. Clearly the amount of quantization noise generated by an A/D converter depends on the size of
the quantization interval. More quantization bits translate into a narrower quantization interval and, hence,
into a lower amount of quantization noise.
Figure 6.6: Quantization Levels
In Figure 6.6, the spacing ∆ between two consecutive quantization levels corresponds to one least sig-
nicant bit (LSB). Usually, it is assumed that quantization noise is signal-independent and is uniformly
distributed over 0.5 LSB and 0.5 LSB. Figure 6.7 also shows the quantization noise of an analog signal
quantized by a 3-bit A/D converter and the corresponding bit stream.
Figure 6.7: Quantization of an Analog Signal by a 3-Bit A/D Converter: (a) Output Signal and
Quantization Error, (b) Histogram of Quantization Error, (c) Bit Stream
147
6.1.3 A/D and D/A Conversions

Because it is not possible to have an actual analog signal within a computer programming environment, an
analog sinusoidal signal is often simulated by sampling it at a very high sampling frequency. Consider the
following analog sine wave:
x (t) = cos (2π 1000t) (6.2)
Sample this sine wave at 40 kHz to generate 0.125 seconds of x (t). Note that the sampling interval, seconds,
is very short, so x (t) appears as an analog signal.
Sampling involves taking samples from an analog signal every seconds. The above example generates a
discrete signal x [n] by taking one sample from the analog signal every seconds. To get a digital signal, apply
quantization to the discrete signal.
According to the Nyquist theorem, an analog signal z can be reconstructed from its samples by using the
following equation:
∞
X t − kTs
z (t) = z [kTs ] sinc (6.3)
Ts
k=−∞
This reconstruction is based on the summations of shifted sinc (sinx/x) functions. Figure 6.8 illustrates the
reconstruction of a sine wave from its samples achieved in this manner.
Figure 6.8: Reconstruction of an Analog Sine Wave Based on its Samples, f = 125 Hz and fs = 1 kHz
It is dicult to generate sinc functions by electronic circuitry. That is why, in practice, one uses an
approximation of a sinc function. Figure 6.9 shows an approximation of a sinc function by a pulse, which
is easy to realize in electronic circuitry. In fact, the well-known sample and hold circuit performs this
approximation.
Figure 6.9: Approximation of a Sinc Function by a Pulse
6.1.4 DTFT and DFT

Fourier transformation pairs for analog and discrete signals are expressed in Table 6.1. Note that the
discrete-time Fourier transform (DTFT) for discrete-time signals is the counterpart to the continuous-time
Fourier transform (CTFT) for continuous-time signals. Also, the discrete Fourier transform (DFT) is the
counterpart to the Fourier series (FS) for continuous-time signals as shown in Table 6.2 . Table 6.3 shows a
list of these transformations and their behavior in the time and frequency domains.
x (t) e−jω0 kt dt
1
R T /2
Xk = −T /2
T
Fourier series for periodic analog signals { , where T denotes
x (t) = k=−∞ Xk ejω0 kt
P∞
period andω 0 fundamental frequency
N nk
PN −1 −j 2π
X [k] = n=0 x [n] e , k = 0, 1, ..., N − 1
Discrete Fourier transform (DFT) for periodic dis- {
N nk
1
P N −1 j 2π
crete signals x [n] = N k=0 X [k] e , n = 0, 1, ..., N − 1
Table 6.2: Fourier series pairs for analog and digital signals
Time domain Spectrum characteristics Transformation type

Continuous (periodic) Discrete FS
Continuous (aperiodic) Continuous CTFT
Discrete (periodic) Discrete (periodic) DFT
Discrete (aperiodic) Continuous (periodic) DTFT
Table 6.3: Dierent Transformations for Continuous and Discrete Signals

149
2
6.2 Lab 6: Analog-to-Digital Conversion, DTFT and DFT
6.2.1 Sampling, Aliasing, Quantization and Reconstruction

The example in this section addresses sampling, quantization, aliasing and signal reconstruction concepts.
Figure 6.10 shows the completed block diagram of this example, where the following four control parameters
are linked to a LabVIEW MathScript node:
Amplitude to control the amplitude of an input sine wave
Phase to control the phase of the input signal
Frequency to control the frequency of the input signal
Sampling frequency to control the sampling rate of the corresponding discrete signal
Number of quantization levels to control the number of quantization levels of the corresponding digital
signal
To simulate the analog signal via a .m le, consider a very small value of time increment dt (dt = 0.001).
To create a discrete signal, sample the analog signal at a rate controlled by the sampling frequency. To
simulate the analog signal, use the textual statement xa=sin(2*pi*f*t), where t is a vector with increment
dt = 0.001. To simulate the discrete signal, use the textual statement xd=sin(2*pi*f*n), where n is a vector
with increment dn. The ratio dn/dt indicates the number of samples skipped during the sampling process.
Again, the ratio of analog frequency to sampling frequency is known as digital or normalized frequency. To
convert the discrete signal into a digital one, perform quantization using the LabVIEW MathScript function
round. Set the number of quantization levels as a control.
To reconstruct the analog signal from the digital one, use a linear interpolation technique via the Lab-
VIEW MathScript function interp1. The samples skipped during the sampling process can be recovered
after the interpolation. Finally, display the Original signal and the Reconstructed signal in the same graph
using the functions Build Waveform, Merge Signal and Waveform Graph. Discrete waveform, Digital wave-
form, Analog frequency, Digital frequency and Number of samples skipped in ADC are also included in the
front panel, shown in Figure 6.11. Use this VI to examine proper signal sampling and reconstruction.

Figure 6.10: Block Diagram of Sampling, Aliasing, Quantization and Reconstruction

151
Figure 6.11: Front Panel of Sampling, Aliasing, Quantization and Reconstruction

6.2.1.1 Analog and Digital Frequency

2πf
Digital frequency ( θ) is related to analog frequency ( f) via the sampling frequency, that is, θ =
fs .
Therefore, one can choose the sampling frequency ( fs ) to increase the digital or normalized frequency of an
analog signal by lowering the number of samples.
6.2.1.2 Aliasing
Set the sampling frequency to fs = 100Hz and change the analog frequency of the signal. Observe the
output for fs = 10Hz and fs = 210Hz (See Figure 6.12 and Figure 6.13). The analog signals appear entirely
dierent in these two cases but the discrete signals are similar. For the second case, the sampling frequency
is less than twice that of the analog signal frequency. This violates the Nyquist sampling rate leading to
aliasing, which means one does not know from which analog signal the digital signal is created. Note the
value of digital frequency is 0.1 radians for the rst case and 2.1 radians for the second case. To prevent any
aliasing, keep the digital frequency less than 0.5 radians.
153
Figure 6.12: Analog and Discrete Waveforms with fs = 100 Hz and f = 10 Hz

155
Figure 6.13: Analog and Discrete Waveforms with fs = 100 Hz and f = 210 Hz
6.2.1.3 Quantization
Now change the Number of quantization levels for some xed values of Frequency and Sampling Frequency.
As the number of quantization levels is increased, the Digital waveform becomes smoother and a smaller
amount of quantization error or noise is generated.
157
Figure 6.14: Digital Waveform with Dierent Numbers of Quantization Levels: (a) 8, (b) 6, (c) 32
6.2.1.4 Signal Reconstruction

Next, set the frequency f = 100Hz and vary the sampling frequency. Observe the reconstructed waveform.
Figure 6.15 shows the reconstructed signals for three dierent values of skipped samples. If the sampling
frequency is increased, fewer samples are skipped during the analog-to-digital conversion, which makes the
reconstruction process more accurate.
159
Figure 6.15: Signal Reconstruction with Dierent Number of Samples Skipped in ADC: (a) 20, (b)10,
(c) 5
6.2.2 DTFT and DFT

In this example, let us compute and compare the DTFT and DFT of digital signals with the CTFT and
FS of analog signals. Figure 6.16 illustrates the completed block diagram of this transform comparison
system. As discussed previously, to simulate an analog signal, consider a small time interval (dt = 0.001).
The corresponding discrete signal is considered to be the same signal with a larger time interval (dt1= 0.01).
161
Figure 6.16: Block Diagram of a DTFT and DFT Transformation System
Generate a periodic square wave with the time period T = 0.1. Connect the input variable mode to an
Enum Control to make the signal periodic or aperiodic. If the signal is periodic (case 0), compute the FS
of the analog signal and the DFT of the digital signal using the fft function over one period of the signal.
For aperiodic signals, only one period of the square wave is considered and the remaining portion is padded
with zeros. For aperiodic signals, the transformations are CTFT (for analog signals) and DTFT (for digital
signals), which are computed using the fft function. In fact, this function provides a computationally ecient
implementation of the DFT transformation for periodic discrete-time signals. However, because simulated
analog signals are actually discrete with a small time interval, this function is also used to compute the
Fourier series for continuous-time signals. Because DFT requires periodicity, one needs to treat aperiodic
signals as periodic with a period T = ∞to apply this useful function. That is why the fft function is also
used for aperiodic signals to compute CTFT and DTFT (as done in the earlier labs). However, in practice, it
should be noted that the period of the zero padded signal is not innite but assumed long enough to obtain
a close approximation. Apply the same approach to the computation of CTFT and DTFT. Because DTFT
is periodic in the frequency domain, for digital signals, repeat the frequency representation using the textual
statement yd=repmat(yd,1,9), noting that the fft function computes the transformation for one period
only.
163
Figure 6.17: Front Panel of a DTFT and DFT Transformation System: Aperiodic Signal
Figure 6.18: Front Panel of a DTFT and DFT Transformation System: Periodic Signal
Figure 6.17 and Figure 6.18 illustrates the front panel of the above transformation system. It shows
the Analog signal and Discrete signal in the time and frequency domains using two waveform graphs. The
transformation type is also shown in the front panel for both of the signals.
165
6.2.3 Telephone Signal

Now let us examine a DFT application. In a touch-tone dialing system, the pressing of each button generates
a unique set of two-tone signals, called dual-tone multi-frequency (DTMF) signals. A telephone central oce
processes these signals to identify the number a user presses. The tone frequency assignments for touch-tone
dialing are shown in Figure 6.19.
Figure 6.19: Frequency Assignments for Touch-Tone Dialing
The sound heard when a key is pressed is a signal composed of two sine waves. That is
x (t) = sin (2πf 1 t) + sin (2πf 2 t) (6.4)
For example, when a caller presses 1, the corresponding signal is
x1 (t) = sin (2π 697t) + sin (2π 1209t) (6.5)
Other than touch-tone signals, modern telephone systems use DTMF event signals for dial tone, busy tone
and ringing tone. Table 1 lists the frequency and timing for standard DTMF event signals.
Tone type Frequency Timing

Dial tone 350 and 440 Hz Continuous
Ringing tone 480 and 620 Hz Repeating cycles of 2 s on, 4 s o
Busy tone 480 and 620 Hz 0.5 s on, 0.5 s o

Table 6.4: DTMF Event Signals
In this application, let us examine the touch-tone dialing system of a digital telephone. Ten input
variables (k0 , k1 , ....., k9 )are assigned to the telephone keys (0, 1, ....., 9). Each input is connected to a Boolean
OK
control. Dierent types of Boolean controls can be created on the front panel. For this application, use
Buttons(Controls → Modern Boolean → OK), each of which is marked with a number from 0 to
→
9. The selected operation properties of an OK Button is Switch until released, as shown in Figure 6.20.
This ensures that the corresponding signal gets generated when a key is pressed and gets back into its initial
position when the key is released.
Figure 6.20: Operation Properties of an OK Button

167
k10 , to act as a counter to count the number of times the keys are pressed. At
Use another input variable,
the beginning, when no key is pressed, the value of k10 is zero and the system returns the dial tone (350 and
440 Hz continuous tone). When the value of k10 is equal to 10, meaning that the keys were pressed for a total
of 10 times, the system assumes that a valid phone number is dialed and returns the busy tone or ringing
(k0 , k1 , ....., k9 ) to a Compound Arithmatic function (Functions
tone. Connect all the Boolean inputs
→ Programming → Boolean → Compound Arithmatic) and select the OR mode. The output of this
function becomes true(1) if any number key is pressed. The result is connected to the Case selector input of
a case structure. The input variable is also connected to the case structure. For true case, k10 is connected to
an Increment function (Functions→Programming →Numeric →Increment) and for the false case,
While Loop(Functions →Programming
it is kept unchanged. The entire system is wrapped inside a
→Structures →While Loop). The output of the case structure is then connected to a Greater or Equal
function (Functions →Programming →Comparison →Greater or Equal) to ensure that the program
exits from the while loop when is greater than or equal to 10. The system shows the ringing tone or busy
tone in the graph and plays the waveform. Figure 6.21 shows the completed block diagram of the touch-tone
telephone system.
Figure 6.21: Block Diagram of Touch-Tone Telephone System

169
Figure 6.22 shows the front panel of the touch-tone telephone system. When the program is run, one can
hear the dial tone and see the signal displayed in the upper waveform graph. As soon as any number key is
pressed, the dial tone is stopped and the corresponding key pad tone is heard and displayed. When keys are
pressed 10 times (a valid phone number), the system plays the ringing tone or busy tone depending on the
setting and displays the tone in the lower waveform graph.
Figure 6.22: Front Panel of a Touch-Tone Telephone System

171
6.2.4 Lab Exercises

Dithering
Dithering is a method of decreasing the distortion of a low-frequency signal due to signal dig-
itization [12]. Dithering works best when the sample rate is high in comparison with the rate at
which the signal changes.
To see how this works, consider a slowly varying signal and its digitization, shown in Fig-
ure 6.23(a). If noise is added to the original signal amplitude roughly at one half the step size, the
signal will look like Figure 6.23(b). If the digitized signal is passed through a resistor-capacitor
circuit to smooth it out, an approximation to the original signal can be recovered. There is no
theoretical limit to the accuracy possible with this method as long as the sampling rate is high
enough.
Design a system to analyze the dithering technique. First, show the digitized and smoothed
signal without dithering. Then, add random noise to the input signal (noise level should not exceed
50 percent of the step size of the digitized signal) and show the digitized and smoothed version.
Measure the maximum and average error between the original signal and recovered signal.
Figure 6.23: Processing at One Half-Step Size: (a) From Top, the Original, Digitized and Smoothed
Signal without Dithering, (b) From Top, the Noise Added, Digitized and Smoothed Signal with Dithering

Image Processing
DFT is widely used in image processing for edge detection. A digital image is a two-dimensional
signal that can get stored and processed as a two-dimensional (2D) array. In the frequency domain,
with the center denoting (0,0) frequency, the center portion of this 2D array contains the low-
frequency components of the 2D signal or image. The edges in the image can be extracted by
removing the low-frequency components.
Read and display the image le image1.jpg provided on the book website. Then, complete the
following steps:
1. Compute and display the 2D DFT of the image using the LabVIEW MathScript functions
fft2 and fftshift.
2. Remove the low-frequency components of the image. A user-controlled threshold can be
specied to remove a varying amount of the low-frequency components.
3. Compute and display the inverse 2D DFT of the image using the LabVIEW MathScript
functions ifft2 and fftshift. The processed image should reect the edges in the original
image.

DTMF Decoder
Design a decoder VI for the DTMF system described in Telephone Signal section. The VI should
be capable of reading the touchtone signal as its input and display the corresponding decoded key
number as its output.
173
Solutions to Lab 6: Analog-to-Digital Conversion, DTFT and DFT

Chapter 7
Analysis of Analog and Digital Systems
1
7.1 Analysis of Analog and Digital Systems
In the previous labs, dierent mathematical transformation tools to represent analog and discrete signals
were examined. This nal lab builds on the knowledge gained in the previous labs to show how to use these
tools to perform signal processing.
7.1.1 Analog Filtering

Analog lters are dened over a continuous range of frequencies. Four basic kinds of analog lters are lowpass,
highpass, bandpass and bandstop. Figure 7.1 shows the ideal characteristics of these lters. In the noise
removal example of Lab 5 , an ideal lowpass lter was used to remove high-frequency noise. However, the ideal
characteristics are not physically realizable and actual lters can only approximate the ideal characteristics.
The RC series circuit analyzed in Lab 3 and Lab 4 is a simple example of an analog lowpass lter.
175
176 CHAPTER 7. ANALYSIS OF ANALOG AND DIGITAL SYSTEMS
Figure 7.1: Characteristics of Ideal Filters (a) Lowpass, (b) Highpass, (c) Bandpass, (d) Bandstop
The voltage output for the circuit shown in Figure 7.2 is given by [9]:
1/ (jωC)
Vout = Vin (7.1)
R + 1/ (jωC)
Figure 7.2: RC Series Circuit Used as Analog Lowpass Filter

177
The magnitude and phase response can be easily found to be [9]:
Vout 1
| H (ω) |=| |= √ (7.2)
Vin 1 + ω 2 R2 C 2

Vout
∠H (ω) = arg = −tan−1 (ω RC) (7.3)
Vin
If the positions of R and C are interchanged, a simple analog highpass lter is obtained as shown in
Figure 7.3.
Figure 7.3: RC Series Circuit Used as Analog Highpass Filter
The voltage output for this circuit is given by
R
Vout = Vin (7.4)
R + 1/ (jωC)
The corresponding magnitude and phase responses are
Vout 1
| H (ω) |=| |= p (7.5)
Vin 1 + 1/ (ω 2 R2 C 2 )

Vout −1 1
∠H (ω) = arg = tan (7.6)
Vin ω RC
7.1.2 Digital Filtering

Digital signal ltering is a fundamental concept in digital signal processing. Two basic kinds of digital lters
that are widely used are FIR and IIR:
FIR (nite impulse response) lters having nite unit sample responses
IIR (innite impulse response) lters having innite unit sample responses
Unit sample response denotes the output in response to a unit input signal. It is common to express
digital lters in the form of dierence equations. In this form, an FIR lter is expressed as
N
X
y [n] = bk x [n − k] (7.7)
k=0
where b's denote the lter coecients and the lter order. As described by this equation, an FIR lter uses
a current input x [n] and a number of previous inputs x [n − k] to generate a current output y [n].
The dierence equation of an IIR lter is given by
N
X M
X
y [n] = bk x [n − k] − ak y [n − k] (7.8)
k=0 k=1
where b's and a's denote the lter coecients and N and M the number of zeros and poles, respectively.
As indicated by Equation (8), an IIR lter uses a number of previous outputs y [n − k] as well as a current
and a number of previous inputs to generate a current output y [n].
In general, as compared to IIR lters, FIR lters require less precision and are computationally more
stable. Table 1 lists some of the dierences between FIR and IIR lters. For the theoretical details on these
dierences, refer to [12].
Attribute FIR lter IIR lter

Stability Always stable Conditionally stable
Computational complexity More operations Fewer operations
Precision Less coecient precision required Higher coecient precision required
Table 7.1: FIR Filter Attributes versus IIR Filter Attributes
where b's and a's denote the lter coecients and and the number of zeros and poles, respectively. As
indicated by Equation (8), an IIR lter uses a number of previous outputs y [n − k] as well as a current and
a number of previous inputs to generate a current output y [n].
b0 + b1 e−jω + ... + bN e−jNω

H (ω) = (7.9)
1 + a1 e−jω + ... + aM e−jMω
It is well-known that a direct-form implementation expressed by Equation (9) is sensitive, in terms of
stability, to coecient quantization errors. Noting that the second-order cascade form produces a more
response that is more resistant to quantization noise [10], the above transfer function is often written and
implemented as follows:
Ns
Y b0k + b1k e−jω + b2k e−j2ω
H (ω) = (7.10)
1 + a1k e−jω + a2k e−j2ω
k=1
where Ns = bN/2c and b.c represents the largest integer less than or equal to the argument. This serial or
cascaded structure is illustrated in Figure 7.4.
Figure 7.4: Cascaded Filter Stages
Each second-order lter is considered to be of direct-form II, shown in Figure 7.5, for its memory eciency.
One can implement each second-order lter in software. Normally, digital lters are implemented in software,
but one can also implement them in hardware by using digital circuit adders and shifters.
179
Figure 7.5: Second Order Direct-Form II

2
7.2 Lab 7: System Response, Analog and Digital Filters
7.2.1 Response of Discrete-Time Systems

This lab involves analyzing the response of discrete-time systems. Responses are calculated for three dierent
kinds of inputs; impulse, step and sine. Figure 7.6 shows the completed block diagram. Connect the input
variable w to an Enum Control so that an input type (impulse, step or sine) can be selected. The response
of this system to any discrete-time input x [n] can be written as
X X
y [n] = bi x [n − i] + ai y [n − i] (7.11)
i i
For this example, consider ve b's and four a's. The system output is displayed using a waveform graph.

181
Figure 7.6: Block Diagram of a Discrete-Time System
Figure 7.7 shows the front panel of the above system. The front panel can be used to interactively select
the input type and set the coecients a and b. The system response for a particular type of input (impulse,
step or sine) is shown in the waveform graph.
Figure 7.7: Front Panel of a Discrete-Time System
7.2.2 Square Root

As another example of discrete-time systems, let us consider taking the square root of an integer number.
Often computers and calculators compute the square root of a positive number A using the following recursive
equation:

1 x [n]
y [n] = y [n − 1] + (7.12)
2 y [n − 1]
If the input x [n] to this equation is set as a step function of amplitude A, then y [n] converges to the square
root of A after several iterations.
183
Figure 7.8 shows the block diagram for a square root computation system. The number of iterations
required to converge to the true value is shown in the output. The initial condition Initial value is set as a
control. Figure 7.9 shows the corresponding front panel.
Figure 7.8: Block Diagram of a Square Root Computation System

Figure 7.9: Front Panel of a Square Root Computation System
7.2.3 Analog and Digital Filtering

In this section, let us examine a basic analog and digital ltering example by implementing a lowpass and a
highpass lter in the analog and digital domains, respectively. Figure 7.10 shows the completed block diagram
of the ltering system. For analog approximation of the signals, use a higher sampling rate (dw1=0.01). To
detect whether the ltering is lowpass or highpass, use the Enum Control Analog lter type. Calculate the
magnitude and phase response of these lters using equations provided in Chapter 7 for analog and digital
lters. Set the values of R and C as controls, and display the responses using a Build Waveform function
and a waveform graph.
185
Figure 7.10: Block Diagram of an Analog and Digital Filtering System

For the digital case, use a lower sampling rate (dw2=0.001). With the Enum controls Digital lter type 1
and Digital lter type 2, select lowpass or highpass and FIR or IIR lter type. Use a Build Waveform function
and a waveform graph to display the magnitude and phase responses of the digital lters. Figure 7.11 shows
the front panel of this ltering system. For a better view of magnitude response of the digital lter, set the
properties of the waveform graph as shown in Figure 7.12 .
Figure 7.11: Front Panel of an Analog and Digital Filtering System

187
Figure 7.12: Graph properties of magnitude response of digital lter
7.2.4 Lab Exercises

Bandpass and Bandstop Filters
Use the lowpass and highpass lters (both analog and digital) described in Analog and Digital
Filtering section to construct bandpass and bandstop lters. The bandpass lter should be able to
pass signals from 50 to 200 Hz and the bandstop lter should be able to stop signals from 150 to
400 Hz. Determine the values of R and C required for this analog lter design. Also, determine the
values of the coecients required for an equivalent IIR digital lter design.

Noise Reduction
Use an analog lowpass lter to remove the high-frequency noise described in Noise Reduction
example of Lab 5. Repeat using a digital lowpass lter.

Frequency Division Multiplexing (FDM)
FDM is widely used in digital communication to simultaneously transmit multiple signals over
a single wideband channel (for details, refer to [14]). For FDM communication, individual signals
are multiplied with dierent carriers to avoid overlaps in the frequency domain. Their time do-
main processing and corresponding frequency spectrums are shown in Figure 7.13. Build a VI to
implement an FDM communication system for three signals x1 (t) , x2 (t)and x3 (t). Use the les
echo_1.wav and retrucksiren.wav on the book website and a random noise with a frequency range
of 20 Hz to 20 kHz to serve as these signals.
Figure 7.13: FDM Communication System

FDM Detector
Build a VI to implement an FDM detector system for detecting the signal x1 (t) as shown in
Figure 7.14.
189
Figure 7.14: FDM Detector

Solutions to Lab 7: System Response, Analog and Digital Filters

Chapter 8
1
References
191
192 BIBLIOGRAPHY
Bibliography
[1] D. Benson. Music: A Mathematical Oering. Cambridge University Press, rst edition edition, 2006.
[2] R. Bishop. LabVIEW 8 Student Edition. Prentice Hall, 2006.
[3] O. Ersoy. Diraction, Fourier Optics and Imaging. Wiley, 2006.
[4] National Instruments. Getting Started with LabVIEW. National Instruments,

http://www.ni.com/pdf/manuals/323427a.pdf, 2003.
[5] National Instruments. LabVIEW User Manual. National Instruments,

http://www.ni.com/pdf/manuals/320999e.pdf, 2003.
[6] National Instruments. Developing Algorithms Using LabVIEW MathScript: Part 1 8211; The LabVIEW
MathScript Node. National Instruments, http://zone.ni.com/devzone/cda/tut/p/id/3256, 2007.
[7] National Instruments. Inside LabVIEW MathScript. National Instruments,

http://zone.ni.com/devzone/cda/tut/p/id/3502, 2007.
[8] B. Heck; E. Kamen. Fundamentals of Signals and Systems Using the Web and MATLAB. Prentice-Hall,
third edition edition, 2006.
[9] S. Karris. Signals and Systems with MATLAB Applications. Orchard Publications, second edition
edition, 2003.
[10] N. Kehtarnavaz. Digital Signal Processing System-Level Design Using LabVIEW. Elsevier, second
edition edition, 2008.
[11] B. Lathi. Linear Systems and Signals. Oxford University Press, second edition edition, 2004.
[12] S. Mitra. Digital Signal Processing: A Computer Based Approach. McGraw-Hill, second edition edition,
2000.
[13] C. Phillips; E. Riskin; J. Parr. Signals, Systems and Transformations. Prentice-Hall, third edition
edition, 2002.
[14] J. Proakis. Digital Communications. McGraw-Hill, fourth edition edition, 2000.
[15] M. Roberts. Signals and Systems. McGraw-Hill, rst edition edition, 2004.
[16] M. Schetzen. Airborne Doppler Radar: Applications, Theory, and Philosophy. AIAA Publisher, rst
[17] J. Buck; M. Daniel; A. Singer. Computer Explorations in Signal and Sytems Using MATLAB. Prentice-
Hall, second edition edition, 1996.
[18] S. Soliman; M. Srinath. Continuous and Discrete Signals and Systems. Prentice-Hall, second edition
edition, 1998.
193
194 BIBLIOGRAPHY
[19] D. Fannin; R. Ziemer; W. Tranter. Signals and Systems: Continuous and Discrete. Prentice-Hall, fourth
[20] YAMAHA. DX7 Digital Programmable Algorithm Synthesizer. YAMAHA Manual, 1999.
INDEX 195
Index of Keywords and Terms
Keywords are listed by the section with that keyword (page numbers are in parentheses). Keywords
do not necessarily appear in the text of the page. They are merely associated with that section. Ex.
apples, 1.1 (1) Terms are referenced by the page they appear on. Ex. apples, 1
1 1. Introduction to LabVIEW, 1 Compound Arithmatic, 167

Containers pallete, 15
2 2. Introduction to MathScript, 1
Controls, 18
3 3. Convolution and Linear Time-Invariant

Create, 22
Systems, 1
D Data Type, 25, 25
4 4. Fourier Series and Its Applications, 1

data.mat, 43
Decorations pallete, 15
5 5. Continuous-Time Fourier Transform and Its Delta, 67
Applications, 2 Diraction of Light, 138

Dithering, 171
6 6. Digital Signals and Their Transforms, 2 Divide, 20, 20, 22, 22
7 7. Analysis of Analog and Digital Systems, 2

Doppler Eect, 102, 137
DTMF Decoder, 172
A Add, 20, 20, 22, 22, 22 Duplicate Page, 15
Add Input, 62
Add Page After, 15
E Echo Cancellation, 81
edit item, 54
Add Page Before, 15
election, 8
Add Terminal, 23
Enum Control, 54, 54, 54, 89, 111, 180
Appearance, 25
Attribute, 178 F FDM Detector, 188
Au, 8 FIR lter, 178
Automatic Tool Selection, 18 for loop, 13, 14, 40
aveform, 111 Frequency, 165
average.m, 41 Frequency Division Multiplexing (FDM), 188
B Bandpass and Bandstop Filters, 187

Frequency domain, 109
Boolean control, 129 G G, 111

break, 40 Graph Indicators palette, 9
Build Waveform, 44, 50, 54, 89, 111, 129 Graphical Programming, 1.1(4)
Build XY Graph, 45, 45 Greater or Equal, 167
Bundle, 129 Greater or Equal?, 26
Bundle by Name, 15
Buttons & Switches palette, 8 H hile, 167
Byte, 25
I if, 40
C case structure, 14 IIR lter, 178
Change to Control, 22 Image Processing, 172
Choose Data Type, 51, 63 Impulse Noise Reduction Using Median
Circuit Analysis, 136 Filtering, 83
Clean Up Wire, 21 Increment, 167
Command Window, 37
196 INDEX
K Knobs, 25 S S, 8
Save, 22, 29
L L, 167
Snapshot, 29
LabVIEW, 1.1(4)
Spectrum characteristics, 148
LabVIEW MathScript function, 89
Start, 29
LEDs palette, 9
Stop, 27, 27, 29
Logical Operators, 41
String Control, 129
M Memory Usage, 29 Synthesis of Electronic Music, 103
Morse Coding, 136

T T, 8
MSE, 67
Tab Control, 15, 73, 89
N Noise Reduction, 188

Text Controls palette, 8
Text Indicators palette, 9
Noise Reduction Using Mean Filtering, 83
Time domain, 109, 148
Numeric, 18
Timing, 165
Numeric 2, 18
Timing Details, 29
Numeric Constant, 20
Timing Statistics, 29
Numeric Controls palette, 8
tomatic, 8
Numeric Indicators palette, 9
Tone type, 165
O OK Button, 166 Tools, 18
OK Buttons, 166 Tools Palette, 18
ool, 8 Transformation type, 148
oop, 167
U Unbundle by Name, 15
P palette, 18, 18
V View As Icon, 21
Patterns, 23
Visible Items, 43
Play, 131
Play Waveform, 81, 81, 81, 102, 129, 137
W W, 111, 167
Probe, 28
Waveform Graph, 50, 54, 129
Prole Memory Usage, 29
Waveform type, 89
Properties, 25, 54, 109
while, 40
R raph, 111
while loop, 14, 26, 26, 27
www.utdallas.edu/∼kehtar/signals-systems, 2
Relational Operators, 41
Remove Page, 15
X x, 37
Remove Terminal, 23
x*.y, 37
Representation, 25
x./y, 37
RL Circuit Analysis, 101
x.^p, 37
Run, 20, 23
XY graph, 45, 129
Run Continuously, 113
Run script, 41, 41 Y y, 37
ATTRIBUTIONS 197
Attributions
Collection: An Interactive Approach to Signals and Systems Laboratory

Edited by: Nasser Kehtarnavaz, Philipos Loizou, Mohammad Rahman
URL: http://cnx.org/content/col10667/1.14/
License: http://creativecommons.org/licenses/by/3.0/
Module: "Preface"
By: Nasser Kehtarnavaz, Philipos Loizou, Mohammad Rahman
URL: http://cnx.org/content/m19826/1.1/
Pages: 1-2
Copyright: Nasser Kehtarnavaz, Philipos Loizou, Mohammad Rahman
Module: "LabVIEW Programming Environment"

Pages: 4-16
Module: "Lab 1: Introduction to LabVIEW"

Pages: 17-30
Module: "LabVIEW MathScript and Hybrid Programming"

Pages: 34-36
Module: "Lab 2: LabVIEW MathScript and Hybrid Programming"

Pages: 37-56
Module: "Convolution and Linear Time-Invariant Systems"

Pages: 59-62
198 ATTRIBUTIONS
Module: "Lab 3: Convolution and Its Applications"

Pages: 62-84
Module: "Fourier Series"

Pages: 87-88
Module: "Lab 4: Fourier Series and Its Applications"

Pages: 89-106
Module: "Continuous-Time Fourier Transform"

Pages: 109-110
Module: "Lab 5: CTFT and Its Applications"

Pages: 111-139
Module: "Digital Signals and Their Transforms"

Pages: 141-148
Module: "Lab 6: Analog-to-Digital Conversion, DTFT and DFT"

Pages: 149-172
Module: "Analysis of Analog and Digital Systems"

Pages: 175-179
ATTRIBUTIONS 199
Module: "Lab 7: System Response, Analog and Digital Filters"

Pages: 180-189
Module: "References"
By: Mohammad Rahman
Page: 191
Copyright: Mohammad Rahman
About Connexions
Since 1999, Connexions has been pioneering a global system where anyone can create course materials and
make them fully accessible and easily reusable free of charge. We are a Web-based authoring, teaching and
learning environment open to anyone interested in education, including students, teachers, professors and
lifelong learners. We connect ideas and facilitate educational communities.
Connexions's modular, interactive courses are in use worldwide by universities, community colleges, K-12
schools, distance learners, and lifelong learners. Connexions materials are in many languages, including
English, Spanish, Chinese, Japanese, Italian, Vietnamese, French, Portuguese, and Thai. Connexions is part
of an exciting new information distribution system that allows for Print on Demand Books. Connexions
has partnered with innovative on-demand publisher QOOP to accelerate the delivery of printed course
materials and textbooks into classrooms worldwide at lower prices than traditional academic publishers.

PDF PDF

Uploaded by

Copyright:

Available Formats

PDF PDF

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

PDF PDF

Uploaded by

Copyright:

Available Formats

An Interactive Approach to Signals and

Rice University, Houston, Texas

2 LabVIEW MathScript and Hybrid Programming

3 Convolution and Linear Time-Invariant Systems

5 Continuous-Time Fourier Transform

6 Digital Signals and Their Transforms

7 Analysis of Analog and Digital Systems

LabVIEW Programming Environment

1.1.1 Virtual Instruments (VIs)

1.1.1.1 Front Panel and Block Diagram

1 This content is available online at <http://cnx.org/content/m19828/1.6/>.

Figure 1.1: LabVIEW Windows: Front Panel and Block Diagram

1.1.1.2 Icon and Connector Pane

1.1.2 Graphical Environment

Figure 1.2: Functions Palette

1.1.2.2 Controls Palette

Figure 1.3: Controls Palette

1.1.2.3 Tools Palette

Figure 1.4: Tools Palette

1.1.3 Building a Front Panel

Figure 1.5: Control Palettes

Figure 1.6: Indicator Palettes

1.1.3.3 Align, Distribute and Resize Objects

Figure 1.7: Menu to Align, Distribute, Resize and Reorder Objects

1.1.4 Building a Block Diagram

Figure 1.9: Icon versus Expandable Node

1.1.4.2 Terminal Icons

Figure 1.10: Terminal Icon Examples Displayed in a Block Diagram

Figure 1.11: Basic Wire Types

1.1.4.4.1 For Loop

Figure 1.12: For Loop

1.1.4.4.2 While Loop

Figure 1.13: While Loop

1.1.4.4.3 Case Structure

Figure 1.14: Case Structure

1.1.5 Grouping Data: Array and Cluster

1.1.6 Debugging and Proling VIs

1.1.6.2 Prole Tool

1.1.7 Containers and Decoration Tools

Figure 1.15: Tab Control

Figure 1.16: Starting LabVIEW

1.2.1 Building a Simple VI

1.2.1.1 Sum and Average VI Example Using Graphical Programming

2 This content is available online at <http://cnx.org/content/m19837/1.5/>.

Figure 1.17: Blank VI

Figure 1.18: Front Panel Conguration

Figure 1.20: Wiring Block Diagram Objects.

Figure 1.21: Completed Block Diagram.

Figure 1.22: VI Verication

1.2.2 SubVI Creation

Figure 1.24: Editing SubVI Icon.

1.2.3 Using Structures and SubVIs

Locate a Greater or Equal? function from Functions →Programming →Comparison →Greater

Figure 1.27: While Loop Enclosure.

Figure 1.28: Front Panel as VI Runs.

1.2.4 Debugging VIs: Probe Tool

Figure 1.29: Probe Tool.

1.2.5 Prole Tool

1.2.6 Lab Exercises

1.1.6 Debugging and Proling VIs

1.1.6.2 Prole Tool

Figure 1.18: Front Panel Conguration

Figure 1.22: VI Verication

1.2.5 Prole Tool

Table 2.1: Some common special characters used in .m les

sum nds sum of the elements of a vector

nd nds indices of nonzero

Table 2.2: Some common functions used in .m les