WiMAX RF Tuning Guide 20090122 a 1.0

Product Name Confidentiality

WiMAX RNP INTERNAL

Product VersionTotal pages: 43

1.3

WiMAX RF Tuning Guide

(For Internal Use Only)

Prepared by Zhang Chao, Zhang Mao Date 2008-11-20

Reviewed by Date

Reviewed by Date

Approved by Date

Huawei Technologies Co., Ltd

All rights reserved

WiMAX RF Tuning Guide INTERNAL

Revision Record

Date Version Description Reviewed by Author

2008-11-14 1.00 First Version.

Hong Weifeng,

Wang Hui, Wang

Yibing

Zhang Chao,

Zhang Mao

2023-04-09 Huawei Confidential Page 2 of 43


Contents

1 Overview....................................................................................................................................... 8

2 Work Flow of RF Optimization....................................................................................................8

3 Test Preparations...................................................................................................................... 10

About This Chapter.................................................................................................................10

3.1 Setting the Objectives.......................................................................................................10

3.2 Dividing Clusters...............................................................................................................11

3.3 Determining the Test Route..............................................................................................12

3.4 Preparing Tools and Documents.......................................................................................12

3.4.1 Preparing Software.................................................................................................12

3.4.2 Preparing Hardware................................................................................................13

3.4.3 Preparing for Documents........................................................................................13

4 Data Collection.......................................................................................................................... 14


4.1 Collecting the Data............................................................................................................14

4.2 Checking the GW Configuration Data...............................................................................15

4.3 DT..................................................................................................................................... 15

4.4 CQT and Indoor Test........................................................................................................17

4.4.1 CQT........................................................................................................................ 17

4.4.2 Indoor Test.............................................................................................................18

5 Coverage Analysis..................................................................................................................... 19


5.1 Coverage Problems and Common Measures...................................................................20

5.1.1 Poor coverage........................................................................................................20

5.1.2 Intra-Frequency Interference Caused by Cross-Cell Coverage..............................20

5.1.3 Unbalance of Uplink and Downlink.........................................................................25

5.2 Coverage Analysis Flow....................................................................................................26

5.2.1 Downlink Coverage Analysis..................................................................................26

5.2.2 Intra-Frequency Interference Between Sites..........................................................29

6 Interference Analysis................................................................................................................33


6.1 Definition and Effect of Interference..................................................................................33

6.2 Internal Interference..........................................................................................................35

6.2.1 Interference Caused by the Asynchronization........................................................35

6.3 External Interference.........................................................................................................37

6.3.1 Effects and Analysis method..................................................................................37

6.3.2 Solution................................................................................................................... 39



7 Handover Analysis....................................................................................................................40


7.1 Optimization of Neighbor cell List......................................................................................40

7.1.1 Deletion of Redundant Neighbor cells....................................................................40

7.2 Settings of Handover Threshold........................................................................................41

7.3 Optimization of Handover Delay........................................................................................41

8 Adjustment................................................................................................................................. 42

9 Summary.................................................................................................................................... 43

10 Attachment................................................................................................................................ 43



List of Tables

Table 3-1 List of WiMAX RF optimization objectives...............................................................10

Table 3-2 List of recommended software for RF optimization.................................................12

Table 3-3 List of recommended hardware for RF optimization................................................13

Table 3-4 Documents need to be collected before optimization..............................................13

Table 4-1 Items of configuration parameter check..................................................................15



List of Figures

Figure 2-1 Flow of RF optimization...........................................................................................9

Figure 3-1 Division of clusters in a project..............................................................................11

Figure 4-1 Traced signaling collected on the LMT..................................................................17

Figure 4-2 Distribution of indoor test points............................................................................19

Figure 5-1 Scanning result of intra-frequency interfering sectors by the scanner...................23

Figure 5-2 Interference caused by improper settings of antenna downtilt angle....................24

Figure 5-3 Location of the interference point in Monument Park............................................30

Figure 5-4 Terminals of the interfered area in Monument Park and result of frequency

scanning by the scanner..................................................................................................30

Figure 5-5 RSSI coverage in Monument Park........................................................................31

Figure 5-6 INR distribution of the interfered area when only Sector 77 is enabled.................31

Figure 5-7 CINR distribution of the interfered area when only Sector 76 is enabled..............32

Figure 5-8 CINR distribution of the interfered area when Sectors 76 and 77 are enabled.....32

Figure 5-9 CINR distribution of the interfered area after the optimization...............................33

Figure 6-1 Components of interfering and interfered communication systems.......................34

Figure 6-2 Spectrum diagram for the analyzed data collected from the RRU of an interfered

sector............................................................................................................................... 39

Figure 7-1 Synchronization checking of the site clock through scanner.................................42



WiMAX RF Optimization Guideline

Keywords

WiMAX, network optimization, RF optimization

Abstract

This document describes how to perform RF optimization. It includes the objectives,

flow, steps, and input and output of RF optimization, as well as the precautions in RF

optimization.

List of acronyms

Acronym and Abbreviation Full Spelling

CINR Carrier-to-Interference-and-Noise Ratio

DT Drive Test

KPI Key Performance Indicator

MS Mobile Station

BS Base Station

RF Radio Frequency

GW Gate Way

RTWP Received Total Wideband Power

RSSI receive signal strength indicator

VIP Very Important People

BBU Base Band Unit

MIMO multiple input multiple output

RRU Radio Remote Unit

AAS Adaptive Antenna System

AMC adaptive modulation and coding

ANT Antenna



1 Overview

With the number of users increasing, the system performance has deteriorated. RF

optimization is necessary, which covers the following main tasks:

Optimizing the signal coverage:

1) Reduce the poor coverage area to ensure that signals can cover the area

continuously.

2) Control the coverage of each cell and overlap area.

Removing interference: There are two kinds of interference, internal interference

and external interference. Internal interference is caused by asynchronous clock

or some other reasons of the system. External interference is inter-system

interference, which is caused by misuse of frequencies. Interference will lead the

degradation of CINR, hard to access network and decrease the network

capacity.

Optimizing the handover process:

1) Check all parameter of the cell are setting correctly.

2) Optimize engineering parameters to control the handover area.

The organization of this document is:

Chapter 1 describes the objectives and scope of work of RF optimization

Chapter 2 describes the work flow and the process of RF optimization.

Chapter 3–8 describes the test preparations, data collection, analysis, and

measurement method according to the RF optimization flow.

Chapter 9 summarizes the document.

Chapter 10 describes some outstanding problems as example.

2 Work Flow of RF Optimization

After all sites are installed and verified, RF optimization could start. The purpose of

the RF optimization is to control the interference and the handover. Once the KPI

requirements have been satisfied, RF optimization is completed.

The work flow of RF Optimization is shown in Figure 1-1. The data collection, problem

analysis, and adjustment should be repeated until the network quality meets the KPI

requirements.



RF optimization starts

Test preparationsEstablish the optimization objectivesDivide clustersDetermine the test pathPrepare for tools and documents

DTCQT (including indoor test)BTS configuration data collection

Adjust engineering parametersConfigure adjacent cell parameters

Problem analysisCoverage analysisInterference analysisHandover analysis

Whether RF indexes meet the KPI requirements?

RF optimization ends

Y

N

Data collection Adjustment measures

···

·

··

··

····

Figure 1-1 Flow of RF optimization

Test preparations: Divide the clusters reasonably. Determine the drive test route with

customers. Prepare for the tools and documents to guarantee the RF test can be

done properly.

Data collection: Collect the data by DT, CQT (including indoor test), and signaling

trace, which is used for problem analysis.

Data analysis: Check the coverage, interference and handover by data analysis, and

propose the adjustment suggestion to customers. With customer permission, the

suggestion is applied. Then, the above job will be repeated circularly until all KPI

requirements are satisfied.

Some problems such as downlink interference, access failure, and network

disconnection are caused by weak signal strength, interference, or missing neighbor.

After RF optimization, output the list of updated engineering parameters and cell

parameters.



3 Test Preparations

About This Chapter

Section Describes

3.1 Setting the objectives The objectives of RF optimization.

3.2 Dividing Clusters The rules for division of clusters.

3.3 Determining the Test route The rules for selecting the drive test route.

3.4 Preparing Tools and

Documents

The tools and documents required for RF

optimization.

3.1 Setting the Objectives

RF optimization focuses on solving the problems of coverage hole, interference, and

handover failure. While different operations have different KPI requirements, KPI

definitions. The purpose of RF optimization is to meet the KPI requirements defined in

the contract.

In general, the network should meet the KPI requirements listed in Table 1-1 after RF

optimization. This KPI list is used only a reference for RF optimization goals. The KPI

should be defined according to the contract, which is an agreement with operators .

Note: The values of CINR and RSSI vary with terminal model. Pls confirm with the

customer about the terminal model for acceptance test.

Table 1-1 List of WiMAX RF optimization objectives

Acceptance

ItemReference Value Remarks

CINR ≥ 10 dB

Urban area ≥ 95% The network should be 0 loading. In the

planned coverage area, the test route is

grid-shaped, and passes through all

sectors coverage area. Suburban area ≥ 90%

RSSI ≥ –85 dBm

Urban area ≥ 95% The network should be 0 loading. In the

planned coverage area, the test route is

grid-shaped, and passes through all

sectors coverage area. (If the operator

wants the test result with the penetration

loss, the value of penetration loss should

be added to the RSSI.)

Suburban area ≥ 90%



3.2 Dividing Clusters

RF optimization has been done by a group or a cluster of base stations. The intra-

frequency interference must be taken into consideration during the optimization.

Confirm with the customer about the partition of clusters, and consider the following

factors:

The number of clusters should be determined according to the actual situation. It

is recommended that 10–25 sites for a cluster.

A cluster should not cross over areas covering different services.

The cluster partition of the other system could be a reference.

Effect of terrain: Terrain can affect the transmission of signal. A mountain may

obstruct the signal transmission. Therefore, a mountain could be as a natural

boundary for clusters. Water could make wireless signals to transmit farther.

Therefore,

If the river is narrow, the sites at both sides could be put one cluster when

transportation conditions permit.

If the river is broad, the river could be a boundary.

In general, it is more typical to divide clusters as a cellular shape than a bar

shape.

Divide clusters based on administrative regions: When the network covers

several administrative regions, the clusters can be divided based on different

administrative regions.

DT workload: Try to make sure that drive tests of each cluster can be completed

in one day. In general, one drive test lasts about four hours.

Figure 1-2 shows an example of dividing clusters in a project. In the following figure,

JB03 and JB04 are dense urban areas; JB01 is an expressway area; JB02, JB05,

JB06, and JB07 are common urban areas; JB08 is a suburban area. Each cluster

contains about 18–22 sites.

Figure 1-2 Example of dividing clusters in a project



3.3 Determining the Test Route

Pls get an agreement with the customer about the DT route before doing drive test. If

the customer determines a DT route, the DT route should include it. If it cannot be

fully adopted due to some reasons, pls explain it to the customer immediately.

The DT acceptance route is the core of the test route of RF optimization. The DT

acceptance route is a key route. Subsequent tasks such as parameter optimization

and acceptance test are all based on the DT KPI route. The test route selection

according to the following rules:

(2) The drive test route should cover main streets, important places, and VIP/VIC.

(3) The test route should cover all cells; involve at least two tests (preliminary test

and final test). Perform tests for all streets in RF planned area if we have time.

(4) To observe the performance change, each drive test should follow same route.

(5) Consider one-way street, traffic light restriction, and turning restriction.

Note that the politics, economy, culture, and living habits of some overseas countries

are different from those in China. Therefore, certain regions may be forbidden to

access. Therefore, before determining the test route, pls communicate with local

drivers, local people and customer.

3.4 Preparing Tools and Documents

Before RF optimization, pls install necessary software (see Table 1-1 in 3.4.1 ),

hardware (see Table 1-2 in 3.4.2 ), and documents (see Table 1-3 in 3.4.3 ), which

is helpful for test analysis. For details, see the following tables.

3.4.1 Preparing Software

Table 1-1 List of recommended software for RF optimization

Number Software Functions Remarks

1 Genex Probe Drive test V2.1 or later; mandatory

2 XCAL-X Drive testV3.1.5.30 or later;

optional

3 Genex AssistantDT data analysis,

neighbor cell checkV2.1 or later; mandatory

4 XCAP-X DT data analysis Optional

5 M2000 Performance analysis Mandatory

6Mapinfo/

Google Earth

Map display and route

selectionMandatory



See the matching between DT software and terminals in the attachment.

3.4.2 Preparing Hardware

Table 1-2 List of recommended hardware for RF optimization

Number Device Item Remarks

1 SCANNER

Frequency scanning and

interference analysis;

optional

2Test terminals

and data cableGCT, CPE, and so on Mandatory

3 LaptopPM1.7G/1G/20G/USB/

COM/PRNMandatory

4 GPS GPS Mandatory

5Car-carried

power inverter

DC to AC, higher than

300 WMandatory

6 Software licenseLicense of PROBE and

ASSISTANTMandatory

7 Hardware dongleDongle of XCAP-X and

XCAL-XOptional

8Spectrum

analyzer

Used to find external

interference; optional

9 Compass Optional

10 Camera Optional

3.4.3 Preparing for Documents

Table 1-3 Documents need to be collected before optimization

Number DocumentsWhether Is

NecessaryRemarks

1 Engineering parameter table Yes

2 Map Yes Mapinfo or paper map

3 KPI requirement Yes

4 Network configuration parameters Yes



Number DocumentsWhether Is

NecessaryRemarks

5 Survey report No

6Checklist of verification at a single

siteNo

7 Plan of the floor to be tested Yes For indoor test

4 Data Collection

About This Chapter

Section Describes

4.1 Collecting the Data Certain methods of data collection and test orders in

RF optimization.

4.2 Checking the GW

Configuration Data

The effect of collecting the GW configuration data

on RF optimization.

4.3 DT The contents, methods, and precautions of DT.

4.4 CQT and Indoor TestThe methods and precautions for data collection

during the CQT and indoor test.

4.1 Collecting the Data

RF optimization focuses on the strength of wireless signals on the network. Test

methods mainly include DT test and CQT (including indoor test). Before the test, pls

get the test devices and determine the test route. In addition, cooperate with other

engineers to check the gateway, BS work properly. If any alarm is reported, clear the

alarm before performing the test. Make record during the test, and record reported

alarms, which helps judge whether problems found in the test (if any) are related to

alarms.

After all preparations are completed, we do DT test and CQT (including indoor test).

Collect the wireless signal data received by scanner or terminal through the DT test.

Analyze the test data for outdoor signal coverage, handover, and interference.

In the GCW network, CQT test actually means the call quality test. In the WiMAX

network, data services are the major services. Therefore, the CQT test for the WiMAX



network is different from that of the GCW network. CQT focuses on the service quality

for fixed points. Discovering, analyzing, and solving RF problems could be done by

CQT.

Indoor test focuses on indoor coverage (such as building, shopping mall, and metro),

hot area (stadium and governmental organ), and tested areas required by the

operator (such as VIC and VIP). Indoor test can also be used to optimize the

handover between indoor places, and between indoor and outdoor.

4.2 Checking the GW Configuration Data

During RF optimization, collect the data of neighbors and the parameters configured

in the GW database. Check the configuration currently is consistent with that

(planning data) before or not. Check the alarm information.

When checking the configuration, the incorrect data, parameters or any alarm are

reported product support engineers. RF engineers do DT, CQT, or indoor coverage

test after the problem is solved

When checking the configuration data, pls focus on handover parameters and power

control parameters. For details, see Table 1-4. For handover parameters, mainly

check the neighbor cell list.

Table 1-4 Items of configuration parameter checking

Class Item

Handover

The function of LST NBS for neighbor cells is implemented on

WASN LMT. (For the 2.1 and 3.2 versions, this item can be

queried directly on the base station.)

Power setting

Whether the function of display carrier-basic-info is enabled.

Query the information at a single site in telnet mode. (This

operation is required in the 2.1 version. In the 3.2 version, this item

can be queried directly on the LMT.)

Cell

Configuration

Whether the function of display carrier-status is enabled in a cell.

Query the information at a single site in telnet mode. (This

operation is required in the 2.1 version. In the 3.2 version, this item

can be queried directly on the LMT.)

After checking, output the updated Wireless Parameter Configuration Data Table and

the parameter modification record, which are used for problem analysis and

subsequent optimization. For operations on WASN LMT and BTS LMT, pls see the

operation guideline in the WiMAX – O&M center Operation Guideline.



4.3 DT

DT includes:

Test for network access status through MSs without loading. This test focuses on

the network access successful rate and the coverage area.

Test FTP download or upload by MSs: This test focuses on the access success

rate, and the average download and upload rate of the FTP server.

Test http webpage open during network access by MSs: This test focuses on the

setup successful rate and the setup delay of the http protocol.

Test ping operation during network access through MSs: This test focuses on the

delay and packet loss rate of the ping operation.

After determining the test route, checking the data configured on the GW and the

alarms, perform the tests for corresponding tasks. Pay attention to the following

points:

(6) Whether all devices are ready before departure.

(7) Whether the computer power is sufficient.

(8) Whether the cigarette lighter in car can provide power properly.

(9) Whether the car-carried power inverter functions properly.

(10) Whether the GPS has sufficient power, and is configured correctly.

(11) Whether all software is installed on the computer, and the licenses are in trial

period.

Confirm that all preparations are correct before departure.

Perform corresponding tests according to the work plans, and record the names of

different test files according to different test contents. In this way, the problem can be

easily located and analyzed.

When finding a problem during the test, create a log file to save it. Perform several

tests for problematic areas, and collect the data as much as possible.

For detailed test methods and settings, pls refer to the drive test guide.

(Optional) Enable the signaling trace function during the DT. Analyze the drive test

data with the traced message to locate the problem. Log in to the LMT and enable the

signaling trace function by either of the following methods:

(12) Log in to the LMT remotely through a terminal, and trace the signaling.

(13) Trace the signaling on the LMT of the GW or within a site with the help of

customer service engineers.



The traced signaling should correspond to the name of the DT data. In this way, the

corresponding traced signaling can be located easily without searching for it in a large

quantity of traced data. For example, if the log file is named Pedestrian Street

20080101 during the DT, the file should also be named Pedestrian Street 20080101

during the signaling trace.

During the signaling trace, collect the following data:

R1, R6, and R3 trace signaling messages of a single subscriber

Real-time feature monitoring records (measured values of RSSI and CINR)

Figure 1-1 Traced signaling data collected on the LMT

Analyze the signaling data accordingly after it is collected. For details on traced data

and collection methods, see the LMT operation guide.



4.4 CQT and Indoor Test

4.4.1 CQT

CQT means the call quality test, that is, the pointing test. The test items for the

WiMAX network are different from traditional voice services because of the features

of the WiMAX network. For the WiMAX network, the test focuses on data services

even though it also involves voice test for VOIP-enabled areas. Select certain

important locations and places for testing.

The information about test locations should be recorded in detail. For example, the

longitude and latitude of the test locations, as well as CINR and RSSI values of the

test locations. In the case of special areas, record the information about their

surrounding, or take photos for the areas.

Test items are as follows:

Test for network access through MS: Perform 100 access tests (the number of

tests may be different, depending on local requirements), and measure the

number of successful tests. This test can be performed through the autocall

function of probe.

Test for FTP upload and download through MS: Perform 10 tests (the number of

tests may be different, depending on local requirements), and measure the peak

value and average value of the throughput, as well as the number of failed tests.

This test can be performed through the autocall function of probe.

Test for http webpage open through MS: Perform 20 tests (the number of tests

may be different, depending on local requirements), and measure the time taken

to open a webpage, and the number of failed tests. This test can be performed

through the autocall function of probe.

Test for ping operation through MS: Perform 20 times (the number of tests may

be different, depending on local requirements), and measure the maximum delay

and average delay, as well as the number of failed tests. This test can be

performed through the autocall function of probe.

For detailed test methods, see the methods in the WiMAX-DT&CQT Test Guide.

CQT must be performed at some typical places (test points), with the security taken

into full consideration. When performing the CQT overseas, pay special attention to

the security of CQT test points. Be sure to communicate with local employees about

various local taboos and insecure areas.



4.4.2 Indoor Test

There are several drive test tools. Take the XCAL-X tool as an example to introduce

the methods of indoor test, as shown below. For detailed operation methods and

settings, see specific tool guides.

When performing an indoor test, add test points manually on the map to display the

moving route because GPS signals cannot be received during the indoor test.

To open the indoor test map, choose File > Inbuilding, or click the Inbuilding icon in

the toolbar. Perform the test according to the following steps:

First, import the plan of the building to be tested through the key. If no plan is available, add test points on the grid map directly.

Then, add test points on the grid map through the key. Add totally eight test

points, as shown in the following figure. The length of the test route is displayed

at the upper right corner of the window.

Figure 1-2 Distribution of indoor test points

Finally, click the key at test point 1, and then move to test point 2. When

reaching test point 2, click the key. In this way, the tool can record the tested data automatically, and associate the data with the moving route. The

information about the tested data and the moving route can be displayed in the

subsequent analysis and playback.

Repeat the previous step until the entire route containing all test points is tested.

Save the log data.

For detailed test methods, see the corresponding test guide.

Like CQT, pay more attention to the security when performing the indoor test.



5 Coverage Analysis

About This Chapter

Section Describes

5.1 Coverage Problems and

Common Measures

Common problems and related solutions for poor

coverage, cross-cell coverage, and unbalance of

uplink and downlink.

5.2 Coverage Analysis FlowThe flow and methods of coverage analysis, and

detailed cases.

Coverage analysis is the major task of RF optimization. Coverage analysis focuses

on signal distribution. Poor coverage, cross-cell coverage, and unbalance of uplink

and downlink are the objects of coverage analysis.

5.1 Coverage Problems and Common Measures

5.1.1 Poor coverage

I. Definition and Effect of Poor coverage

Poor coverage means the RSSI of signals in covered areas is less than –85 dBm,

and the CINR of signals is less than 10 dB. At present, manufacturers have no unified

standards for producing terminals. Thus, the values of RSSI and CINR measured by

various terminals are different. To claim a poor coverage area, identify the type of

terminals used locally, perform feasibility tests on site, and compare the signal

strength in the tested area with the baseline value.

Valley, back of mountains, elevator well, tunnel, underground garage, basement, and

inside of tall buildings are often areas with poor coverage. In these areas, the values

of RSSI and CINR are very low. That is, RSSI < –85 dBm, and CINR < 10 dB. Too

low CINR cannot meet the minimum requirement for network access. Thus, the

successful rate of network access in these areas is low, or network disconnection

occurs immediately after successful access.

II. Optimization Measures for Poor coverage

In general, the following measures are taken to optimize the poor coverage:

Optimize the coverage by increasing the carrier frequency power, adjusting

antenna azimuth, downtilt angle, or height, or using the antenna with a higher

gain.

Adjust the antenna at terminal side, move the terminal closer to the window, or

use the terminal with directional and high gain antenna.



Replace the indoor type terminal with the outdoor type one, and adjust the

antenna to ensure the quality of received signals.

Install the RRU on the tower so that the loss generated by feeders is less and

strong coverage is achieved.

Adjust the times of repeated coding.

Build new base stations in the case the poor coverage is caused by

discontinuous coverage. In this case, build an overlapped coverage area that is

large enough. In addition, pay attention to the interference resulted from wider

coverage areas.

Build new base stations in the areas with poor coverage that is caused by valley

and back of mountains so that the signals can cover wider areas.

5.1.2 Intra-Frequency Interference Caused by Cross-Cell Coverage

I. Analysis on Possible Causes

In the ideal situation, the signals of each cell should be controlled strictly and should

not cover the area outside the designed range. The wireless environment, however, is

complicated. For example, the terrain, building distribution, street distribution, and

water area can affect wireless signals. Therefore, wireless signals cannot be

controlled easily, and the ideal situation cannot be achieved. In this case, cross-cell

coverage occurs.

Only a few frequency resources can be used for the WiMAX network because of the

features of the WiMAX network. To improve the spectrum utilization rate, the

spectrum is reused frequently, which causes intra-frequency interference between

sites easily.

Intra-frequency interference between sites is mainly caused by the interaction of

multiple sites. Therefore, intra-frequency interference between sites mainly occurs in

the urban areas where sites are installed densely, and the areas where the terrain is

complicated. In normal cases, intra-frequency interference between sites happens in

urban areas: tall building, broad street, high tower, crossroad, and water area. Hilly

area and mountain area are the places where the interference occurs easily because

of the terrain.

Other causes for intra-frequency interference between sites are as follows:

(14) Antenna azimuths are irregular

In a network with multiple sites, antenna azimuths should be regular in the entire

network. However, antenna azimuths are irregular because of the signal

coverage, traffic distribution, and restriction on installation locations, which is out

of control. In addition, interference caused by intra-frequency sectors occurs in

certain areas. Therefore, coverage of intra-frequency sectors in one area should

be reduced.

(15) The base station or antenna is located at a high place

If the base station is located in a high altitude, most surrounding areas are in the

line-of-sight range. In this case, signals can be transmitted in a large area. The



overlap area cannot be controlled easily, and intra-frequency interference is out

of control.

(16) The antenna downtilt is set improperly

The antenna downtilt is determined according to the height, the coverage area,

and the antenna model. If the antenna downtilt is set improperly, strong signals

overshoot, which causes interference to other areas. Thus, intra-frequency

interference increases, which will make network disconnection and difficult to

access network.

(17) Antenna performance are not good

The WiMAX network has not been developed for a long time. The performance

of antennae provided by some manufacturers has not been approved. Therefore,

some antenna performance cannot meet the engineering requirement, for

example, the signal strength of both side lobe and back lobe is strong, which can

increase intra-frequency interference to other sites.

II. Analysis on Effect of Intra-Frequency Interference Between Sites

Intra-frequency interference between sites may cause the following network

problems:

(18) Degradation of CINR (some signals are not covered)

Intra-frequency interference will degrade CINR, and decrease the network

throughput.

(19) Difficult access to network

When strong interference exists, the terminal may not access the dominate

sector, or cannot access the network at all.

(20) Reduction of capacity

The coding mode is adjusted for the demodulation threshold of AMC according

to the measured value of CINR, and the interference reduces the value of CINR.

Therefore, the coding mode and the capacity are reduced, which causes lower

average throughput of sectors, and affects the user experience directly.

III. Analysis Methods and Symptom Description of Intra-Frequency Interference Between Sites

(21) By terminals

In general, if RSSI < –70 dBm during the test, the absolute value of the sum of

CINR and RSSI is approximately equal to the absolute value of base station

sensitivity. This rule is only applicable to some terminals. Since the measured

values of CINR for some terminals are too high, the judgment should depend on

the actual situation. If the absolute value of the sum of CINR and RSSI is far less

than the absolute value of base station sensitivity (for example, the difference is

more than 10 dB), unacceptable intra-frequency is between sites. In other words,

if RSSI is in normal value range, but CINR is too low, there is intra-frequency

interference probably.



Simple method to locate the interference source: Disable the sector to which the

terminal belongs, and then do tests using the same frequency. Locate the source

interfering site according to received signals. In addition, observe whether RSSI

and CINR are normal during the test. If RSSI and CINR are abnormal, disable

the active sector, and find other interfering sites. During commercial application,

the active sector cannot be disabled. Therefore, test engineers should judge the

situation on site. This method is not practical, and could be used only before

delivery.

(22) By scanner

A scanner can scan the downlink signals for more than 10 intra-frequency

sectors at the same time; record the values of RSSI and CINR for each intra-

frequency sector, and save the scanning result and GPS data for analysis. To

use this method, the requirements for RF engineer are not high. Therefore, it is

recommended that a scanner be used to check intra-frequency interference.

Scan the areas with possible intra-frequency interference, and then analyze the

scanning result to determine the interference source in the area. As shown in

Figure 1-1, two intra-frequency signals with preamble values of 67 and 68 are

the interference signals in this area. At present, the function of the scanner is not

good enough. RF engineers should propose requirements for scanner after the

trial.

Figure 1-1 Scanning result of intra-frequency interfering sectors by a scanner

IV. Optimization Methods of Intra-Frequency Interference Between Sites

(23) Adjust antenna configuration.



We can change each sector coverage to reduce intra-frequency interference by

adjusting the azimuth and the downtilt. The dominate sector signal strength is

increased, which can reduce the intra-frequency.

To strengthen the signals, engineers can adjust the antenna azimuth so that the

antenna faces to the area directly. To decrease the signals coverage, antenna

azimuth could move from the covered area. The adjustment of antenna downtilt

is similar to that of antenna azimuth. Reduce the antenna downtilt to increase the

coverage, and vice versa.

The adjustment of antenna downtilt should follow certain rules. If the downtilt is

set too small, the cell coverage can be increased, but interference may occur. If

the downtilt is set too large, the cell coverage can be reduced, but the directional

diagram of the antenna may be distorted. In general, the downtilt should not be

more than 6°. If the downtilt is more than 6°, it is recommended that Electric

antenna should be selected.

As shown in Figure 1-1, the interfered area circled by black lines is caused by

unreasonable settings of antenna azimuth and downtilt for Woodlands 2 sector.

The antenna azimuth of Woodlands 2 sector is 310°, and the downtilt is 2°. The

antenna faces directly to the GolfClub site, and covers a large area, which

causes interference directly to the areas near the Golfclub 2 sector.

Change the antenna downtilt of Woodlands 2nd sector from 2° to 7°, and change

the antenna azimuth of Woodlands 2nd sector from 310° to 250°. Thus, the

overshoot coverage of Cell 360 is reduced, and the interference is reduced.

The interference in some areas may hard to be reduced only by antenna

configuration adjustment. In this case, replace the antenna model, add reflecting

device or blocking device, relocate the antenna, or relocate the site if possible.

Figure 1-1 Interference caused by improper settings of antenna downtilt angle

(24) Adjust the transmit power

Although adjusting the antenna azimuth can reduce the interference in some

areas, hotspot areas may not be covered, or new interference comes in other

areas. When the antenna downtilt is increased to a certain degree, the



directional diagram of the antenna may be distorted if the antenna downtilt

continues to increase. Therefore, to reduce the coverage range of some

interfered sectors, reduce the transmit power of this sector. When the antenna

downtilt is reduced to a certain degree, increase the transmit power to increase

the coverage range of the sector. At present, the maximum transmit power of the

site is 40 dBm or 37 dBm. If the transmit power increases beyond the maximum

value, the RRU may be burned out. Therefore, do not increase the transmit

power beyond the maximum value. Power adjustment can be used together with

antenna adjustment.

Note: It is recommended that the transmit power should not be adjusted in RF

optimization. In general, the transmit power is adjusted only when the antenna

system is difficult to adjust.

(25) Adjust the networking mode

At present, the WiMAX16e network adopts three networking modes: PUSC 1/3,

PUSC with all SC 1X3X3, and FFR. The spectrum utilization rates of PUSC 1/3

and PUSC with all SC 1X3X3 are nearly the same. The spectrum utilization rate

of FFR is the highest among the three, but its interference is the strongest. At

present, Huawei provides a networking mode of inter-frequency FFR (for details,

see the related document). This networking mode is an improvement of PUSC

with all SC 1X3X3. That is, the networking mode of PUSC with all SC is used in

the cell center, and the networking mode of inter-frequency FFR is used in the

cell boundary. In this way, the CINR distribution of interfered areas can be

improved effectively.

If the intra-frequency interference cannot be reduced, try to change the

networking mode. The networking mode can be changed in the following order:

FFR -> PUSC 1/3 or PUSC with all SC 1X3X3 -> Inter-frequency FFR. The

change of networking mode, however, requires supports from both the terminal

and the site.

5.1.3 Unbalance of Uplink and Downlink

I. Unbalance of Uplink and Downlink and the Effect

Unbalance of uplink and downlink means that the distance for downlink (uplink)

signals to be identified is more than the transmission distance for uplink (downlink)

signals to be identified. That is, the coverage range is determined by the weakest

signals. In general, the uplink signals of the WiMAX network are restricted. Therefore,

this section only describes the restriction of uplink signals.

In the case of unbalance of uplink and downlink, the terminal can receive signals but

cannot process the services. The terminal can demodulate the signals of the site but

cannot access the network. The rate of network disconnection increases. After the

services are established, the packet error rate increases, and the services jitter

frequently. This can affect the traffic measurement index and user satisfactory level.

Unbalance of uplink and downlink is caused by the following factors:



(26) Different maximum transmit power for the site and the terminal

The transmit power of the RRU ranges from 30 dBm to 40 dBm, depending on

the site type. The transmit power of WiMAX terminals ranges from 23 dBm to 27

dBm because of the radiation, dimension, and battery capacity.

(27) Different sensitivities for the site and the terminal

In general, the receiver sensitivity of the site is higher than that of the terminal,

which can partially offset the restriction of uplink signals.

(28) Disabling of power control or improper setting of power control parameters

When the link losses of the terminal and the site are changed, if the power

control of the site is disabled, the power of the terminal cannot be adjusted

according to channels, which can cause the unbalance of uplink and downlink. If

the power control parameters are set improperly, the power control cannot be

enabled timely, which can also cause the unbalance of uplink and downlink.

II. Solution to the Unbalance of Uplink and Downlink

Use terminals with high power.

Increase the transmit power of terminals.

Replace omni-directional antennae with directional high gain antennae.

Enable the power control, and set the parameters properly. For detailed

parameter values, perform feasibility tests according to the actual situation.

Improve the receiver sensitivity of the site.

5.2 Coverage Analysis Flow

In the actual test, drive test can be used to measure downlink signals only. To

measure uplink signals, perform signal trace on the LMT. In this way, the test point

map cannot be provided in real time, and signals cannot be observed easily.

Therefore, only downlink coverage is analyzed in this document.

5.2.1 Downlink Coverage Analysis

Downlink coverage analysis refers to the analysis of RSSI and CINR measured in the

DT test.

RSSI simply indicates the strength of front-end signals of the receiver. If the strength

of received signals is higher than the threshold, links are in normal state. RSSI can be

estimated and calculated easily, and the received sampling signals are not processed

and demodulated. The received signals contain useful signals, noise, and

interference. Therefore, strong received signals do not mean high channel quality,

and only indicate that the channel contains strong signals.

The quality standard of RSSI should be used together with the optimization standard.

Suppose the optimization standard of RSSI is as follows:

RSSI ≥ –85 dBm ≥ 95%Measurement result of the terminal,

outdoor, no load



The quality standard of RSSI should be defined as follows:

Good: RSSI ≥ –75 dBm

Fair: –85 dBm ≤ RSSI < –75 dBm

Poor: RSSI < –85 dBm

The values of CINR range from –10 dB to 40 dB, depending on terminals.

The quality standard of CINR should be defined as follows:

Good: CINR ≥ 20 dB

Fair: 10 dB ≤ CINR < 20 dB

Poor: CINR < 10 dB

Determine the poor coverage according to the measured values of RSSI for downlink

signals. First, identify the areas with fair coverage and poor coverage for further

analysis. Then, perform analysis for the identified downlink coverage hole areas in

comparison with the measured result of CINR.

If CINR is also poor in the same area, the area is a poor coverage area. Optimize

areas with poor coverage by using different adjustment measures based on different

cases. Perform tests again after the adjustment is completed, and check the

optimization result. If the result meets the requirements, the optimization is

completed. If the result does not meet the requirements, continue to perform the

optimization, test, and analysis, until the problem is solved.

If RSSI is good but CINR is poor, judge whether signals of other sites cross over this

area, which thus causes poor coverage. Disable the sectors of adjacent sites, and

measure the values of RSSI and CINR. If the values are improved, the intra-

frequency interference is caused by cross-cell coverage. If the values are not

improved, take into consideration other interference (for details, see Chapter 6).

If the poor coverage is caused by cross-cell coverage, perform optimization according

to optimization measures for cross-cell coverage. Perform tests after the optimization

is completed, and analyze the test result after optimization. If obvious cross-cell

coverage persists, continue to perform optimization, adjustment, test, and analysis,

until the problem is solved.

I. Poor coverage Case Caused by Intra-Frequency Interference due to Cross-Cell Coverage

Below are analysis and solving processes for the poor coverage case caused by

intra-frequency interference due to cross-cell coverage.

In the project of country P, Huawei adopts the networking mode of 1×3×3 three

frequencies. Four 3-sector sites in a small urban area cover the network continuously.

Perform tests for two subscribers under a same sector of a same site. Trace the two

subscribers on the LMT. Discontinuous error codes are found for uplink signals.

Downlink CINR is fluctuated greatly. Two terminals are disconnected from the

network easily. After network disconnection, the two terminals enter adjacent sectors

sometimes, but return to the original sector easily because of poorer signals.



Problem analysis on the site:

(29) Intra-frequency interference causes poor downlink CINR, and poor downlink

CINR thus causes a high error rate for uplink signals. The uplink signals,

however, are good and poor alternately. This is abnormal because the intra-

frequency interference should be continuous from the viewpoint of time.

(30) The threshold of repeated coding is set too low so that repeated coding cannot

be enabled when signals are poor, which thus causes high error code rate.

(31) External interference causes the fluctuation of uplink and downlink signals. The

signals of another test terminal under the same sector, however, are stable.

(32) The threshold of terminal switching and scanning is set too high. Scanning of

ZyXEL terminals can cause high error code rate for uplink signals easily.

(33) Terminal hardware is faulty so that uplink and downlink signals are not stable.

(34) The output power of the site is fluctuated so that the downlink CINR of the

terminal is fluctuated. The signals of another test terminal, however, are stable.

Final solving process:

(35) Remove intra-frequency interference by disabling adjacent intra-frequency

sectors one by one through remote background maintenance platform. After

adjacent intra-frequency sectors are disabled, observe the values of downlink

CINR of two terminals through the LMT. The values are not improved much.

Therefore, a high error code rate for two subscribers is not caused by intra-

frequency interference from the same network.

(36) Change the threshold of repeated coding for uplink signals from 7 dB to 10 dB.

The error code rate is not improved much.

(37) Perform frequency scanning through spectrum analyzer. No external interference

is found.

(38) Disable the terminal scanning function of either subscriber so that the subscriber

cannot initiate scanning (scanning can cause high error code rate for uplink

signals). The problem persists.

(39) Replace the terminal with a ZyXEL terminal for the subscriber. The problem

persists.

(40) Perform tests with ZyXEL terminal and Huawei terminal for three hours at the

near end of the site. The signals of two terminals are stable. Therefore, the

output power of the site is correct.

(41) Perform tests with two terminals near the location of either subscriber.

Discontinuous error codes are found for uplink signals, and downlink CINR is

fluctuated greatly. Disable intra-frequency sectors of adjacent sites to check

whether the problem is caused by intra-frequency interference from the same

network. After that, uplink and downlink signals are stable, and high error code

rate and network disconnection are not found again. Why the signals are good

and poor alternately in intra-frequency interfering sectors? To answer this

question, perform tests in the interfering sectors, and find that the standing wave

ratio of the antenna system is fluctuated from 1.4 to 2. Thus, the cause for the

problem is that the change of the standing wave ratio causes the strong

fluctuation of transmitting signals of the interfering sectors, which increases the



interference to the interfered sectors accordingly. Why the cause is not located

by removing intra-frequency interference of the networking mode in step 1?

Because intra-frequency interfering sectors are disabled for a very short period,

during which time the interference is weak. The error code rate is low for

interfering subscribers just before and after intra-frequency interfering sectors

are disabled. Therefore, the problem is not caused by intra-frequency

interference from the same network.

(42) Final solution: First, remake feeder connectors for the interfering sectors, and

keep the standing wave ratio in 1.4 stably. Then, reduce the power of interfering

sectors by 3 dB. After that, downlink CINRs of the two subscribers are improved

greatly, but still do not match the RSSIs (RSSI is –60 dBm, and CINR is about 10

dB). The interference is still strong. Finally, add a baffle to the back of the

antenna in interfering sector, which can reduce the intra-frequency interference

caused to other sectors by the back lobe of antenna in the interfering cell. In this

way, the problem is solved. The downlink CINRs of the two subscribers are

stable, and the high-order modulation can be used properly. High error code rate

is not found for uplink signals any more.

5.2.2 Intra-Frequency Interference Between Sites

I. Analysis Flow of Intra-Frequency Interference Between Sites

(43) When performing tests, analyze the values of CINR and RSSI collected through

terminals, and select the area with high RSSI and low CINR as a candidate area

that may involve interference. Or judge the interfered areas according to the

result provided by the scanner.

(44) Judge which adjacent intra-frequency sectors cause the interference in interfered

areas.

(45) Analyze the distribution of RSSI and CINR of related interfered areas, and the

causes for interference according to the actual environment. Provide solutions

for interference according to specific causes. When optimizing the process of

removing interference in an area, take into consideration the effect on other

areas in the network. Removing interference in an area may cause new

interference or coverage hole to other areas.

(46) Perform tests again after the optimization, and analyze the values of RSSI and

CINR for drive test data. If the analysis result cannot meet the KPI requirements,

select new major areas for further optimization until the KPI requirements are

met.

Note that in the new major areas, the cells adjusted in the previous optimization

should not be adjusted again.

II. Example of Optimization for Removing Interference

Take the optimization for removing interference in an office as an example to

introduce the process of interference analysis.

(47) Find interference points



Find the interfering area. This area is designed to be covered by Sector 77, and

is only 400 meters from Sector 77, as shown in Figure 1-1. The drive test shows

that the CINR of the area is very low, and the RSSI is very high. Figure 1-2

shows the result of frequency scanning by the scanner.

Figure 1-1 Location of the interference point in Monument Park

Figure 1-2 Terminals of the interfered area in Monument Park and result of frequency

scanning by the scanner

(48) Analyze the signal distribution of the cells near the interference point



Figure 1-1 RSSI coverage in Monument Park

As shown in Figure 1-1, the average RSSI of this area is about –60 dBm, which is

considered as good coverage.

Figure 1-2 CINR distribution of the interfered area when only Sector 77 is enabled



Figure 1-3 CINR distribution of the interfered area when only Sector 76 is enabled

Figure 1-4 CINR distribution of the interfered area when Sectors 76 and 77 are

enabled

As shown in Figure 1-2 to Figure 1-4, since Sector 77 is designed to cover the

area, the CINR of signals in the area is more than 20 dB when only Sector 77 is

enabled. After Sector 76 is enabled, Sector 77 is interfered greatly by Sector 76

so that the average CINR is reduced to 10 dB or less.

(49) Analysis of data after optimization

The analysis on drive test data and on-site scenarios shows that Sector 76 is

located at a higher place, which thus covers a wider area. Change the downtilt of

Sector 76 to 9°, and the antenna azimuth to 330° so that the interference caused

by Sector 76 to Sector 77 is reduced, as shown in Figure 1-1.



Figure 1-1 CINR distribution of the interfered area after the optimization

The drive test data collected after the optimization shows that the interference

near the interfered area is reduced to a certain degree, and the value of CINR

increases by about 10 dB or more.

6 Interference Analysis

About This Chapter

Section Describes

6.1 Definition and Effect of

Interference

The definition, classification, and effect of

interference.

6.2 Internal InterferenceTwo types of internal interference, effects, and

optimization solutions.

6.3 External InterferenceThe effect and related solution of external

interference.

6.1 Definition and Effect of Interference

Radio interference refers to interference that is generated during the radio

communication, and can reduce the strength of useful signals, damage useful signals,

or block the normal communication. Radio interference signal refers to

electromagnetic energy that enters the channels or systems of receiving device



through direct or indirect coupling. Radio interference signal can affect the receiving

of signals required for radio communication so that the performance is lowered, the

quality is degraded, the information is erroneous or lost, and the communication is

blocked. Therefore, the fact that useless radio signals cause lower quality or damage

of received useful radio signals is called interference. Figure 1-2 shows the

components of typical interfering and interfered communication systems.

Transmit filter

Receive filter

Receiving unit

Receiver

Sending unit

A

B

C

D

Interfering system

Interfered system

Transmitting antenna

Receiving antenna

Figure 1-2 Components of interfering and interfered communication systems

According to interference source, interference can be divided into internal interference

and external interference.

According to the direction of signal transmission links, interference can be divided into

uplink interference and downlink interference. Uplink interference refers to the effect

of received signals due to the site. Downlink interference refers to the effect of

received signals due to the terminal.

According to the frequency band of signals, interference can be divided into inband

interference and outband interference. Inband interference refers to the interference

that is distributed within the receiver bandwidth, and generally can affect the receiver

and the normal running of the system. Outband interference refers to the interference

that is affected by the guard bandwidth, the performance indexes of the receiver filter,

and the outband suppression of the receiver.

The following sections describe the effects caused by the interference to the WiMAX

system, and the solutions/suggestions.



6.2 Internal Interference

The WiMAX 16e system is a TDD broadband communication system that is based on

the OFDMA technology, which features changeable bandwidth and high spectrum

utilization rate. The advantages of the TDD technology are that the spectrum is not

necessarily used in pairs, and the proportion of uplink and downlink subframes is

changeable. The advantages are also the major risks for complicated interference of

the WiMAX 16e system. The WiMAX 16e system focuses on the spectrum utilization

rate, therefore, the system has strict requirements for networking mode, which are

different from the GSM system and the CDMA system. The GSM system has 20 pairs

of frequencies, and can prevent interference effectively through the frequency reuse

technology. The CDMA system has orthogonal spreading codes, and can suppress

interference through the orthogonality of spreading codes when the spectrum

multiplexing level is 1. At present, the technology for removing inter-cell interference

in the OFDMA system is immature. For example, the Interleave-Division Multiple

Access (IDMA) technology is still not applied currently in the WiMAX 16e system

because of the complexity of its algorithm. Therefore, no mature technology for

preventing inter-cell interference can be applied in the WiMAX 16e system currently.

At present, the major problem facing the WiMAX 16e system is internal interference.

Internal interference can be divided into uplink and downlink interference caused by

the asynchronization between sites, and downlink interference caused by intra-

frequency sectors between synchronous sites.

6.2.1 Interference Caused by the Asynchronization

I. Causes and Effects

The WiMAX 16e system is a TDD system, which requires strict synchronization

between uplink and downlink. The reason for the interference caused by the

asynchronization between sites is that the GPS of some sites is unlocked, which

causes asynchronization of signal sending and receiving between sites, and thus

causes serious interference between sites. Asynchronization includes the case where

the time is synchronous with different proportion of uplink and downlink subframes,

and the case where the time is asynchronous. In the case where the time is

asynchronous, BS-BS, SS-SS, BS-SS, and SS-BS interference exists. BS-BS and

BS-SS interference is the strongest. This document describes the BS-BS and BS-SS

interference in the case of asynchronous sites.

For the BS-BS interference, since the site has high transmit power, high receiver

sensitivity, and is located at a higher place, if the GPS of a site is unlocked, which

causes interference between sites, the interference may affect the sites a dozen of

kilometers far from the interfering site. In this case, the uplink RSSI increases, CINR

is degraded, RTWP increases, and service rate is reduced, which can cause failure to

access the network or failure to process services after network access. For the BS-

SS interference, since the CINR of the terminal is currently measured through

preambles. If sites are asynchronous, the preambles of different sites can be sent in



different periods. Therefore, the CINR measured on the terminal is not reduced, but

may be increased. In general, asynchronization between sites does not affect

downlink signals greatly. No such affect is found in the actual test as long as the

terminal can access the network. For the SS-SS interference, since the terminal has

low transmit power, and is used at a lower place, the downlink interference caused by

the terminal to other terminals cannot be measured easily.

II. Symptom Description

In general, in the case of interference caused by the asynchronization between sites,

a subscriber first cannot access to the network easily. On the terminal side, the

downlink CINR is good, but the ranging operation fails, or the network can be

accessed but no uplink service can be processed. On the LMT side of the site, the

uplink RSSI is high but the uplink CINR is low (often from –5 dB to 10 dB), and the

packet error rate of uplink signals is high (kept above 50%). Query the RTWP with no

load received by the RRU through the LMT or RTWP remote query tool (for details on

query methods, see the related guide). The three sectors of the site have high

RTWPs with no load, which are more than –80 dBm on average.

III. Problem confirmation

At present, the interference is located through the RTWP with no load. If the RTWP

with no load of a site is much higher than the normal value, and there are no online

active subscribers, the interference may be caused by the asynchronization between

sites. Here, the normal value means the normal RTWP with no load. Normal values

vary with versions, which range from –90 dBm to –95 dBm in some offices, and from

–100 dBm to 105 dBm in South Africa currently. Why the normal value means the

normal RTWP with no load? Because if a sector has active subscribers, the RTWP of

the sector increases, but generally cannot be more than –85 dBm. To check whether

the interference is caused by the asynchronization between sites, disable the active

sectors of adjacent sites (including the active sectors of this site), and then measure

the RTWP of any remaining enabled sector to judge whether the RTWP is in normal

value range.

Note: At present, scanners produced by Recommend can help observe whether

active sectors are synchronous visually. Therefore, scanners produced by

Recommend are recommended.

IV. Solution

Query the GPS status information and the system clock information (for details on

query methods, see the related guide). If the GPS status information is abnormal

(lacking of time, longitude, and latitude, or having less than four satellites), or the

system clock information is always in the tracking state, check the physical

connection of the GPS, including soft jumper, 1/2 feeder, conversion connector, and

GPS antenna. If any connector is loosened, plug it again tightly. After checking, if the

GPS status is still abnormal, check whether the GPS antenna system is damaged.

Try to replace the GPS antenna, jumper, feeder, and so on. After that, if the GPS



status remains unchanged, try to restart the BBU. Perform rests again to verify the

RTWP after the GPS is in normal state. In general, the process from the restart to

GPS locking consumes 10 minutes.

6.3 External Interference

The frequencies used currently by the WiMAX 16e system are within 2.5G frequency

band and 3.5G frequency band. In principle, the frequencies are seldom interfered by

external factors or other systems. In practice, certain nonstandard systems may

occupy these frequencies, thus causing serious interference to the WiMAX 16e

system.

External interference can be divided into inband interference and outband

interference. Outband interference is caused by insufficient guard bandwidth or less

outband suppression of the receiver.

6.3.1 Effects and Analysis method

External interference can increase the noise floor of the receiver, and thus reduce the

CINR, which at worst can cause the amplifier saturation of the receiver and generate

blocking interference. In this case, the throughput of sectors decreases, the coverage

shrinks, and the network is difficult to access.

Frequency scanning is an effective method to prevent external interference. During

the early stage of network construction, perform frequency scanning in the areas

where sites are to be deployed. Check whether other systems can cause interference

to the frequencies used in the WiMAX 16e system. If yes, request the customer to

clear the interference. This is the best method. At present, the frequency scanning

service in the network planning provided by Huawei is charged. Most carriers do not

want to purchase this service. They believe no external interference occurs to their

networks. In addition, to perform frequency scanning, a special spectrum analyzer is

required. At present, certain carriers do not have a special spectrum analyzer or find it

difficult to share the device (for example, Vodacom in South Africa has only one

spectrum analyzer currently). In addition, this method can only remove the external

interference in the early stage of network construction, rather than that occurs in the

period of network running.

During the running of the WiMAX 16e system, the most direct method to judge

external interference is to query the RTWP with no load of the site. Monitor the RTWP

of each site on a daily basis through RF commissioning station (the V2.1 version only)

or RTWP query tool. In this way, the interference status of each site can be found at

any time. If the RTWP of a sector in a site is high, first check whether the RTWP

increase is caused by active subscribers, and then check whether the RTWP

increase is caused by internal interference resulting from the asynchronization

between sites (for detailed methods, see 6.2.1 “Interference Caused by the

Asynchronization”). If the problem is not caused by internal interference, it is probably

caused by external interference.



In the site with external interference, the RTWP generally has the following features:

(50) The RTWP value is high, but not too high. It generally ranges from –90 dBm to –

75 dBm.

(51) The interference involves direction attribute, that is, only one or two sectors out

of the three have high RTWP values.

(52) The interference involves frequency difference, that is, if a same sector uses

different frequencies, especially uses the f1 and f3 frequencies out of the three,

RTWP values are different.

(53) The interference involves fluctuation attribute, that is, the RTWP values queried

by the RF commissioning station are fluctuated with the time.

(54) The interference involves area attribute, that is, if high RTWP values are found in

the same direction and the same frequency of the sites adjacent to the possible

interfered site, external interference probably occurs. In this case, the direction

and frequency range of the interference source can be judged through the

azimuth and frequency of the interfered sector.

High RTWP values caused possibly by interference can be excluded by the following

methods:

(55) If the main and diversity RTWP values of a same sector are different, and either

of both is normal, external interference is excluded because interference signals

generally do not have the polarization feature. An exception is that high RTWP

values in the form of single polarization occur to a batch of sites.

(56) Try to change the frequencies of the sector where high RTWP values are found.

After many frequencies are changed, if the RTWP values are nearly the same,

external interference is excluded.

Another method to judge external interference is to collect the data of the RF

receiving channel of the RRU in the possible interfered sector, and then submit the

collected data to the headquarters for analysis and simulation. Figure 1-1 shows the

spectrum diagram for the analyzed data collected from the RRU of an interfered

sector in South Africa. As shown in the figure, the sector involves the narrowband

interference caused by inband and outband.



Figure 1-1 Spectrum diagram for the analyzed data collected from the RRU of an

interfered sector

6.3.2 Solution

If external interference probably occurs, request the customer to perform frequency

scanning. Pay attention to the following points during the frequency scanning.

WiMAX is a TDD system. Common scanners cannot obtain synchronous signals from

sites but test the received uplink frames. Therefore, all WiMAX sites adjacent to the

interfered site should be disabled to ensure that the increase of noise floor is not

caused by internal signals.

The interference generally involves direction attribute. Therefore, when performing

frequency scanning, first use an omni-directional antenna to judge whether

interference exists, and then use a directional antenna to determine the direction of

the interference source (if any).

When performing frequency scanning, check the power spectrum of the local

frequency band and adjacent frequency bands. In general, first use broader scanning

bandwidth to find interference easily, and then reduce the scanning bandwidth to

locate the bandwidth and strength of the interference.

If the interference source is located through frequency scanning, report the frequency

band, bandwidth, system duplex mode (TDD system or FDD system), and possibility

of clearance to the customer in a timely manner. In addition, provide an analysis

report on the effect of the interference to the WiMAX system performance, and then

request the customer to clear the interference.



7 Handover Analysis

About This Chapter

Section Describes

7.1 Optimization of Neighbor

cell Common methods of neighbor cell optimization.

7.2 Settings of Handover

Threshold

The effect of handover threshold on handover, and

the optimization methods.

7.3 Optimization of Handover

DelayThe optimization methods for handover delay.

During RF optimization, main handover problems are related to the neighbor cell

optimization and control of handover success rate in the drive test.

The size and location of handover areas can be controlled by adjusting the RF

parameters. In this way, the network disconnection caused by sudden changes of

signals can be reduced, and the handover success rate can increase. At present, the

handover function in the WiMAX system is immature. Huawei products support hard

handover only. In addition, feasibility tests for the handover function are seldom

performed in each office. Therefore, the information about this problem will be added

in the future after the handover function is mature.

7.1 Optimization of Neighbor cell List

Neighbor cell optimization includes neighbor cell adding and neighbor cell deletion. In

the case of missing neighbor cells, the cells with strong signals cannot be added,

which causes increase of interference, and network disconnection. In this case,

necessary cells should be added. In the case of redundant neighbor cells, neighbor

cells have a large quantity of messages, which increase unnecessary signaling

overhead. In addition, necessary cells cannot be added when the neighbor cell list is

full. In this case, redundant neighbor cells should be deleted.

7.1.1 Deletion of Redundant Neighbor cells

In the V2.1 version, the WiMAX system has at most six neighbor cells. If the neighbor

cells reach or exceed six, the necessary missing neighbor cells found during the

optimization cannot be added. In this case, certain redundant neighbor cells should

be deleted. In the V3.2 version, the WiMAX system has at most 32 neighbor cells. In

this version, the preceding method is also applicable.



Take care when deleting redundant neighbor cells. If a necessary neighbor cell is

deleted by mistake, serious results such as network disconnection may occur.

Therefore, the following rules should be followed:

(57) Before deleting an neighbor cell, check the modification history of the neighbor

cell, and ensure that the neighbor cell to be deleted is redundant.

(58) After deleting a redundant neighbor cell, perform all-around tests, including the

drive test and important indoor pointing test. Ensure that no exception occurs.

Otherwise, restore the original data configuration.

During RF optimization, a neighbor cell should be deleted in the following cases:

The neighbor cell with cross-cell coverage should be deleted. The premise is that

problems about cross-cell coverage are solved, and no poor coverage area arises.

A neighbor cell should be deleted according to the network topology and experiences.

This rule is applicable to the case where the original neighbor cell list is full, and a

new neighbor cell needs to be added. Perform tests after deletion, and ensure that

the deleted neighbor cell does not cause bigger problems. Otherwise, reselect the

neighbor cell to be deleted.

During the later running stage, an neighbor cell can be deleted according to the traffic

measurement data.

7.2 Settings of Handover Threshold

Whether the handover threshold is set properly can affect the handover success rate

directly. When the CINR is less than the scanning threshold, the terminal initiates the

scanning. When the CINR reaches the handover threshold, the terminal initiates the

handover. If the scanning threshold is set too low, the terminal is disconnected from

the network easily. If the scanning threshold is set too high, the terminal initiates the

scanning frequently, which wastes air interface resources. If the handover threshold is

set too high, the terminal cannot initiate the handover easily. If the handover threshold

is set too low, ping-pong handover occurs.

Since the CINR values measured for various terminals are different currently, no

unified handover threshold can be set for various terminals. The handover threshold

of the GCT USB dongle is 10 dB, which is calculated based on the feasibility tests in

South Africa. The handover thresholds of other terminals should be determined

through feasibility tests on site.

7.3 Optimization of Handover Delay

The handover delay of terminals is related to the clock synchronization of the system.

If the system has too long handover delay, or the handover success rate is found too

low, try to check whether any problem about synchronization occurs to various sites

of the system. A scanner can help locate the problems about clock synchronization

for different sectors easily. As shown in Figure 1-1, if the clock difference of two

sectors under handover is too large, the site clock should be synchronized.



Figure 1-1 Synchronization checking of the site clock through scanner

8 Adjustment

Besides the neighbor cell list, engineering parameters can also be adjusted in RF

optimization to solve the problem.

Most problems about coverage and interference can be solved through the following

methods:

Adjust the antenna downtilt angle.

Adjust the antenna azimuth.

Adjust the antenna height.

Relocate the antenna.

Change the antenna type.

Add the antenna baffle or install the antenna against the wall.

Relocate the site.

Add a site or RRU.

Relocate the terminal.

Use the terminal with high gain antenna.

Use outdoor type terminal.



9 Summary

(59) No mature standard and method are available for the use of scanner and

terminal together. This section should be provided at any time in the future

development.

(60) At present, the definition of handover area does not follow a unified standard.

How to define the handover area is an outstanding problem. Since not too much

experience can be referred from the legacy network, the definition of handover

area should be added to the RF Optimization Guide in the future when the

concept and application of handover area are mature.

(61) The KPIs and other data in this guide are for reference only. Different offices

may have different test results. Therefore, KPIs and other data should be defined

according to the baseline document. In addition, the measured values are

different from various terminals provided by different manufacturers. Therefore,

KPIs and other data should be defined by engineers after performing feasibility

tests on site.

(62) The indoor test function of the probe has not been developed. This guide

describes only the indoor test method of the couwei. The indoor test method of

the probe can be added to this guide only after this function is implemented for

the probe.

(63) Because of the version limitations, this guide does not describe the problems

about location update and idle status of terminals, which should be added in later

versions.

10 Attachment


WiMAX RF Tuning Guide 20090122 a 1.0

Documents

interference analysis

internal interference

coverage analysis flow

downlink coverage analysis

frequency interference

effect of interference

external interference

zhang mao date date