High RTWP Problem Troubleshooting and Optimization Guide

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18

:

Internal only

Author Vivek Kumar


Staff Number 742914
Tripathi
GSRC India
Radio
Depart Wireless Performance &
Network Product Family:
ment RNP & RNO
Planning
Dept.
Update
2013-12-03 Product Version: V900R013SPH588
Time
Approv
er
Title High RTWP Problem troubleshooting and Optimization Guide
Pheno We were Facing High RTWP problem at many sites in XX network of YY country. High RTWP impacts RF
menon KPIs which resulting in poor customer perception about network.
Descri
ption:
Alarm None
Inform
ation
Cause This document describes process of troubleshooting High RTWP problems. It can be used as technical
Analysi reference for troubleshoot and Optimize poor KPIs caused by high RTWP. If facing high RTWP issue at
s some sites in any network. First of all we should confirm this high RTWP is related with PIM or external
interference or due to high traffic. After knowing the reason behind high RTWP we can take necessary
actions. Detailed method of troubleshooting and optimization solutions are described in this document.

Handli
ng If facing high RTWP issue at some sites in any network. First of all we should confirm this high RTWP is
Process related with PIM or external interference or due to high traffic.

Method to do PIM Testing: Please follow bellow method to confirm if issue is related with PIM or not.

Step 1: Open M2000 and login with user name and password

2017-8-15 1 , 18
:
Internal only

Step 2: M2000 select Main Topology

Step 3: Use CTRL+F to find the site

Step 4: locate site

2017-8-15 2 , 18
:
Internal only

Step 5 select Maintenance Client:

Step 6: Use MML command Lst RRU to know Cabinet, subrack and slot number

Step 7: start the board RTWP by filling cabinet, subrack and slot number

2017-8-15 3 , 18
:
Internal only

Step 8: Open Board RTWP monitoring window

Step 9:Adjust the board RTWP window

Step 10 start the output power monitoring

2017-8-15 4 , 18
:
Internal only

Step 11 Output power monitoring

Step 12 adjust the window as below

2017-8-15 5 , 18
:
Internal only

Step 13 MML: STR RFTEST

Step 14 record screen


Case 1: No PIM interference: if board RTWP not changing with change in output power as shown bellow

Case 2: PIM interference : if board RTWP changing with change in output power as shown bellow

2017-8-15 6 , 18
:
Internal only

Now by this method we can know which cells have high RTWP issue caused by PIM. If we found PIM it
need to be rectified. If NO PIM then we can do FFT testing to confirm if high RTWP caused by some
external interference or not.

Method to do FFT spectrum scan test: FFT spectrum scan is a off-line MML command
So the permission should be approved in advance to do the FFT scan.

Step 1: DEA Cell: we need to deactivate cell as per bellow snap shot

Step 2: Open LMT as per bellow snap shot

2017-8-15 7 , 18
:
Internal only

Step 3: use command LST RRU to know cabinet no, subrack no & slot no

Step 4:SET parameter for FFT as per bellow snap shot

2017-8-15 8 , 18
:
Internal only

Step 5: Save scan result picture

Step 6: ACT ucell

2017-8-15 9 , 18
:
Internal only

Using above FFT spectrum scan method we can see if high RTWP belongs to external interference or not. If
found external interference need to trace the source and turn it off.

Solution for Passive Inter Modulation (PIM): we have found many cells which have PIM issue. We
suggested to solve PIM by finding out PIM location as per bellow methods.

Here is example of one cell (BKF0014_S01) affected by PIM issue. Borad RTWP was increasing with
increase in output power. Hence PIM is confimred.

Locating the PIM interference: Solution 1

2017-8-15 10 , 18
:
Internal only

Locating the PIM interference(solution 2)

If High RTWP issue is not related with PIM & External interference and it is caused by high traffic at site.
Then in this case we can try to do some parameter optimization as suggested bellow.

Solution 1: Optimization of CQI Feedback Period from 2ms to 8 ms


Basic principle and application scenario:
Basic principle: The shorter the CQI feedback period is, the better the downlink data transmission is. The
longer the CQI feedback period is, the lower the load is.
Application scenario: any scenarios where the HSUPA cell capacity needs to be expanded and the cell

2017-8-15 11 , 18
:
Internal only

RTWP needs to be reduced.


Gain and risk
Gain: On the office in country A, the CQI feedback period is changed from 2 ms to 8 ms. After optimization,
the average value of the RTWP traced in real time is reduced by about 8 dB. The average RTWP value of
hour-level traffic statistics is reduced by about 10 dB.
Risk: After this solution is implemented on offices A and B, there is no impact on the entire network.
Version in which the solution is implemented
In RAN13.0, parameters are baselined to 4 ms. Parameters are modified manually in earlier versions

Impact: As in above picture when we modify the CQI feedback period from 2 ms to 8 ms. The RTWP
average value is reduced by about 8 dB in real-time trace (-76.97-> -85.05). The load overshoots in a short
time, and the frequency decreases obviously.

Solution 2 : Access Parameter Optimization

Basic principle and application scenario


Basic principle: By reducing the spike of the preamble open-loop power control on the uplink RTWP in
RACH initial access, the RTWP is significantly reduced in the cell with frequent RACH access (thousands
of times per hour).
Application scenario: This solution is mainly used for indoor coverage scenarios, and for cells with frequent
subscriber access and high RTWP.
Gain and risk
Gain: After the Constantvalue parameter is modified on the office in country C, the RTWP average value is
reduced by 3 dB to 4 dB. After PreambleRetransMax, PowerRampStep and Mmax parameters are modified,
the average RTWP value is reduced by 1 dB.
Risk: For indoor coverage scenarios, the access delay increases.
Version in which the solution is implemented
All

Solution :
For the indoor scenario:
Modify the constant value of the initial transmit power: Constantvalue=-20-> Constantvalue=-30,

2017-8-15 12 , 18
:
Internal only

Modify the Preamble retransmit times: PreambleRetransMax=20 -> PreambleRetransMax=40,


Modify the step: PowerRampStep=2 -> PowerRampStep=1,
Modify the maximum number of cycles: Mmax=8->Mmax=3,
Modify the upper limit of random rollback: NB01max=0 -> NB01max=10.
For the outdoor scenario:
Modify the maximum cyclic times: Mmax=8->Mmax=3,
Modify the upper limit of random rollback: NB01max=0 -> NB01max=10.

Application effect :
After the access parameter optimization, the RTWP average value is reduced by 3 dB to 4 dB.
The RTWP short-time overshoot still exists, but with significantly better frequency and intensity.
The RTWP distribution in the low interval increases significantly.

Solution 3: 10 ms Fixed PO Optimization

Basic principle and application scenario


Basic principle: Configure higher reference PO for the 10 ms PO, to reduce DPCCH SIR working point and
uplink load cost, and to improve the throughput of the HSUPA cell with many subscribers.
Application scenario: This is used for the network with a high ratio of HSUPA 10 ms subscriber, and with
limited uplink capacity.
Gain and risk
Gain: For the lab test with dual antennas, 75% uplink load target threshold, 18 online subscribers and two
uploading subscribers, the gain of the uplink throughput in the cell is 15%.
Risk: The peak rate of the HSUPA 10 ms subscriber in commercial network competition is slightly affected.
Version in which the solution is implemented
All (Versions earlier than R13 involve internal parameter modification. You need to obtain the commands
for modifying internal parameters from the PDT manager of the product line. It is recommended that
HSUPA PO adaptive adjustment is used on R13 version and later versions.)

Solution 4: 2 ms Fixed PO Optimization

Basic principle and application scenario


Basic principle: Configure higher reference PO for the 2 ms PO, to reduce DPCCH SIR working point and

2017-8-15 13 , 18
:
Internal only

uplink load cost, and to improve the throughput of the HSUPA cell with many subscribers.
Application scenario: This is used for the network with a high ratio of HSUPA 2 ms subscriber, and with
limited uplink capacity.
Gain and risk
Gain: For the lab test with one antenna, 90% uplink load target threshold, and with 7 to 8 subscribers
running FTP for upload, the gain of the uplink throughput in the cell is 15%.
Risk: The peak rate of the HSUPA 2 ms subscriber in commercial network competition is greatly affected.
Version in which the solution is implemented
All (If internal parameters are involved, you need to obtain the commands for modifying internal parameters
from the PDT manager of the product line.)

Solution 5: State Transition/EFD

Basic principle
Basic principle: Transit the light-traffic subscribers to the FACH state, reduce dedicated channel resources
occupation of light-traffic subscribers, and reduce uplink and downlink load. EFD is used for networks with
high iPhone penetration rates.
Application scenario: This is used for networks with high penetration rates for intelligent terminals, large
amount of online subscribers and low throughputs.
Gain and risk
Gain: The gain is related to traffic models and the permanent online timer.
Risk: This increases PS call drop rate, and decreases the setup success rate of CS composite services.
Because the EFD is not compatible with other terminals, you are advised to use this solution for iPhones
only.
Version in which the solution is implemented
R12

Application effect :
If the state transition is disabled, the capacity loss is severe, and the whole RTWP increase is obvious (by
about 15 dB).

Solution 6: iPhone Link-Deleting Problem

Basic principle and application scenario

2017-8-15 14 , 18
:
Internal only

Basic principle: When iPhone releases a link, the iPhone retransmits the RRC CONN REL CMP signaling
message. However, the network side releases the link before UE. As a result, the uplink power control is
abnormal and the RTWP spike occurs. The solution is delaying the link release on the network side when
iPhone releases a link.
Application scenario: This is used for networks with high iPhone penetration rates and high RTWP.
Gain and risk
Gain: The RTWP spike is reduced, especially for networks with high iPhone penetration rates. After the
solution is implemented on the site in Thailand, the RTWP average value improves by about 3 dB.
Risk: None
Version in which the solution is implemented
RNC RAN12 SPC516

Application effect:
The RTWP average value increases by about 3 dB (89.9 dBm -> 93.1 dBm).

Solution 7: RTWP Value Optimization of SIB7 Broadcast

Basic principle and application scenario


Basic principle: Reduce the spike of the preamble open-loop power control on the uplink RTWP in RACH
initial access, especially in high load scenarios. The RTWP in SIB7 is broadcast in real time. If you reduce
the RTWP in SIB7, the effect is essentially the same as that of reducing the Constvalue. Both are for
reducing the UE power of transmitting the first preamble.
Application scenario: This is used in the scenario where the RACH causes RTWP ramp.
Gain and risk
Gain: After the RTWP value of SIB7 broadcast is optimized on the office in country A, the average RTWP
value is reduced by about 3 dB.
Risk: In the scenario with low load and not obvious RTWP spike caused by RACH, each RACH access
needs a large amount of preamble ramp progress, which causes hundred-ms-level delay to increase.
Version in which the solution is implemented
NodeB R12SPC430

Application effect:
After the RTWP value of SIB7 broadcast is optimized, the RTWP average value is reduced by about 3 dB:
93.5 vs 90.8.

2017-8-15 15 , 18
:
Internal only

In the real-time trace data, you can see that the frequency of load short-time overshoot obviously becomes
lower.

Solution 8: Adaptive Adjustment of HSUPA PO (power offset)

Basic principle and application scenario


Basic principle: If the subscriber rate is lower than a certain threshold, and the air interface load is limited,
increase the HSUPA reference PO to reduce the SIR working point of lower rate subscribers, and to improve
uplink capacity.
Application scenario: This is used for the network with a high ratio of HSUPA 10 ms subscriber, and with
limited uplink capacity.
Gain and risk
Gain: When the subscriber amount is large and the rate is low, the gain is obvious. For example, more than
fifteen 10 ms subscribers that only have small amounts of data to be retransmitted are online and uploading
at the same time. If the power offset of data channels for these subscribers all can be adjusted, the HSUPA
capacity of the cell can improve by more than 40% without object load increase. This is shown by the
increase of the average throughput or the increase of uplink data transmission subscriber number at the same
time. The effect is better with HSUPA TTI switchover based on the load enabled.
Risk: Four minutes are taken from triggering to completion of this feature. The gain is not obvious for
frequent link-release/link setup subscribers or subscribers with too short lasting time links.
Version in which the solution is implemented
RAN13.0 (fixed PO configuration can be carried out on early versions.)

Solution 9: HSUPA TTI Selection and Switchover Solution Enabled + 2 ms Periodic Retry Disabled

Basic principle and application scenario


Basic principle: If the subscriber rate is lower than a certain threshold, and the air interface resource or the
CE resource is limited, the switchover of the HSUPA subscriber from 2 ms TTI to 10 ms TTI is triggered.
The RTWP overshoot caused by data burst because of high minimum rate of 2 ms subscribers is reduced.
The 2 ms periodic retry is disabled to prevent TTI ping-pong switchover.
Application scenario: This is used for networks with HSUPA 2 ms function enabled, lots of 2 ms subscribers
and limited uplink load.
Gain and risk

2017-8-15 16 , 18
:
Internal only

Gain: The RTWP overshoot caused by data burst because of high minimum rate of 2 ms subscribers is
reduced.
Risk: When the 10 ms TTI subscriber requires high-speed data transmission, the rate rises a little slowly
because of the reconfiguration process to the 2 ms TTI.
Version in which the solution is implemented
RAN10.0

Comparison of Different TTI Switchover Solutions

Solution 10: Reception Using Multiple Antennas

Basic principle and application scenario


Basic principle: For the same transmission rate, the reception using multiple antennas reduces the signal
power that each antenna receives, that is, the uplink load. For the same load, the reception using multiple
antennas allows the UE to send larger transmission blocks, which means the uplink capacity is improved.
Gain and risk
Gain: Compared with the single antenna reception, the dual-antenna reception improves the uplink capacity
by more than 50%. Compared with the dual-antenna reception, the four-antenna reception improves the
uplink capacity by more than 50%.
Risk: No for now.
Version in which the solution is implemented
All

Sugges If facing high RTWP issue use troubleshooting method and optimization techniques mentioned in this
tions document to resolve the issue and improve network performance.
and
summa
ry
Attach
2017-8-15 17 , 18
:
Internal only

ments
Relate
d
docum
ent
link

2017-8-15 18 , 18

You might also like