Gsm bss-network-kpi-handover-success-rate-optimization-manual(HSR)

INTERNAL
GSM BSS Network KPI (Handover Success Rate) Optimization
Manual

Product name

Confidentiality level

GSM BSS

Internal

Product version

Total 30pages

V00R01

GSM BSS Network KPI (Handover
Success Rate) Optimization Manual
For internal use only

Prepared by

GSM&UMTS Network Performance Dong
Research Department
Xuan

Date

2008-4-23

Reviewed by

Date

yyyy-mm-dd

Reviewed by

Date

yyyy-mm-dd

Granted by

Date

yyyy-mm-dd

Huawei Technologies Co., Ltd.
All rights reserved

2012-12-12

华为机密，未经许可不得扩散

第 1 页, 共 30 页

INTERNAL
Manual

Revision Record
Date

Revision
Version

Change Description

Author

2008-4-23

0.88

Draft completed

Dong Xuan

2008-7-30

1.0

Some flowcharts are updated.

Dong Xuan

2012-12-12


第 2 页, 共 30 页

INTERNAL
Manual

Keywords

Handover Success Rate
Abstract

By analyzing the factors that affect the Handover Success Rate (HOSR) on the BSS side, this
document provides a method of quickly locating the cause of low HOSR or slow handover. In
addition, this document provides measures for optimizing the HOSR, thus meeting field engineers'
working requirements for solving handover problems. This document is used for optimizing the
KPIs of network performance and monitoring the network quality.
Acronyms and Abbreviations
Abbreviations

Expansion

AMR
BCCH

Broadcast Control Channel

BER

Bit Error Ratio

BQ

Bad Quality

BSC

Base Station Controller

BSIC

Base Station Identity Code

CDU

Combining and Distribution Unit

CIC

Circuit Identification Code

HOSR

Handover Success Rate

KPI

Key Performance Index

MR

Measure Report

MS

Mobile Station

NE

Network Element

QoS

Quality of Service

RQI

Radio Quality Indication

TA

2012-12-12

Adaptive Multi Rate

Timing Advance


第 3 页, 共 30 页

INTERNAL
Manual

Table of Contents
1 Basic Principles.................................................................................................................8
1.1 Definition....................................................................................................................8
1.2 Theory.......................................................................................................................8
1.3 Recommended Formula............................................................................................8
1.4 Measurement Point...................................................................................................9
2 Influencing Factors..........................................................................................................11
3 Analysis Process and Optimization Method..................................................................11
3.1 Process of Analyzing Handover Problems...............................................................12
3.1.1 General Process of Locating a Handover Problem.......................................12
3.2 Methods for Optimizing Handover Problems...........................................................14
3.2.1 Classification of Handover Problems............................................................14
3.2.2 Hardware and Transmission Failure..............................................................15
3.2.3 Improper Data Configuration.........................................................................17
3.2.4 Congestion of the Target Cell........................................................................19
3.2.5 Clock Problems.............................................................................................21
3.2.6 Interference Problems...................................................................................22
3.2.7 Coverage Problems, and Uplink and Downlink Imbalance............................23
3.2.8 Failed Inter-BSC/Inter-MSC Handovers........................................................24
3.2.9 Automatic Neighboring Cell Optimization......................................................25
3.2.10 Testing Tool Selection and Testing Suggestions..........................................26
3.2.11 Configuration Suggestions for Tests on the Existing Network.....................26
4 Optimization Cases..........................................................................................................27
4.1 A Handover Fails Because the BSIC Cannot Be Decoded......................................27
4.2 A Handover Fails Because Frequency Sequencing of the MS Is Different from That
of the BSC.......................................................................................................................27
4.3 A Handover Fails Due to Unreasonable Parameter Configuration...........................27
4.4 The Number of Failed Incoming BSC Handovers Increases Because the Handover
Request Does Not Contain Class Mark 3........................................................................27
4.5 An Incoming BSC Handover Fails Because the A Interface Phase Flag Is Set
Wrongly...........................................................................................................................28
4.6 Because the Idle Burst Is Enabled, the Interference Increases, the Receiving
Quality Decreases, and the HOSR Becomes Low...........................................................28
4.7 Different HOSRs Resulting from Different Cause Values Contained in the Clear
Command Messages Sent by Different Switches............................................................29
5 Information Feedback......................................................................................................29
5.1 TEMS Log Files About Problem Cells......................................................................29
2012-12-12


第 4 页, 共 30 页

INTERNAL
Manual

5.2 Requirements of Configuration Data of the Existing Network and Traffic
Measurement Feedback..................................................................................................29

2012-12-12


第 5 页, 共 30 页

INTERNAL
Manual

List of Tables
Table 3.1List of handover timers commonly used.............................................................18

2012-12-12


第 6 页, 共 30 页

INTERNAL
Manual

List of Figures
Figure 1.1Signaling process of intra-BSC handover...........................................................9
Figure 1.2Signaling process of inter-BSC handover.........................................................10
Figure 1.3Flow chart of locating a handover problem.......................................................13
Figure 1.4Flow chart and timer description........................................................................19
Figure 1.5Flow chart of automatic neighboring cell optimization....................................26

2012-12-12


第 7 页, 共 30 页

INTERNAL
Manual

GSM BSS Network KPI (Handover Success
Rate) Optimization Manual
1 Basic Principles
1.1 Definition
Handover is an important function in mobile communication systems. As a means of radio link
control, handover enables users to communicate continuously when they traverse different cells.
The HOSR is the ratio of the number of successful handovers to the number of handover requests.
The major purpose of handover is to guarantee call continuity, improve speech quality, reduce
cross interference in the network, and thus provide better services for mobile station (MS)
subscribers.

1.2 Theory
The HOSR is an important KPI of the call hold type. According to the processes, this KPI can be
divided into two types: Handover Success Rate and Radio Handover Success Rate. According to
the relations between involved network elements (NEs), this KPI can be divided into three types:
Success Rate of Intra-BSC Handover, Success Rate of Incoming BSC Handover, and Success Rate
of Outgoing BSC Handover. The HOSR is an important KPI assessed by operators because the
value of the HOSR directly affects the user experience.

1.3 Recommended Formula
The HOSR is obtained through traffic measurement. The recommended formula for calculating
this KPI is as follows:

Handover Success Rate = Successful Handovers/Handover Requests

Radio Handover Success Rate = Successful Handovers/Handover
Commands
For details, refer to the GSM BSS Network KPI (TCH Call Drop Rate) Baseline.

2012-12-12


第 8 页, 共 30 页

INTERNAL
Manual

1.4 Measurement Point
MS

BTS2

BSC

BTS1

MS

MSC

Measurement Report
CHAN ACT

A1

CHAN ACT ACK
Handover Command
B1

Handover
Access
Handover detect
PHY INFO
SABM
EST IND
UA
Handover Complete

Handover Performed
C1

Figure 1.1 Signaling process of intra-BSC handover

2012-12-12


第 9 页, 共 30 页

INTERNAL
Manual
MS

BTS2

BSC2

CHAN ACT

MSC

BSC1

Handover
Required

Handover
Request
A2

BTS1

MS

Me asurement
Re port

A3

CHAN ACT ACK

Handover
Request ACK

Handover
Command

B2

Handover
Access

Handover
Command
B3

Handover
detect
PHY INFO
PHY INFO
SABM
EST IND
UA
Handover Complete
C2
Handover
Complete

Clear
Command
Clear Complete

C3

Figure 1.2 Signaling process of inter-BSC handover

The measurement points illustrated in 1.4 and 1.4 are as follows:
A1——Measurement point of Incoming Internal Inter-Cell Handover Requests and Internal IntraCell Handover Requests
B1——Measurement point of Incoming Internal Inter-Cell Handover Responses (Incoming
Internal Inter-Cell Handovers) and Internal Intra-Cell Handover Commands
C1——Measurement point of Successful Incoming Internal Inter-Cell Handovers and Successful
Internal Intra-Cell Handovers
A2——Incoming External Inter-Cell Handover Requests
B2——Incoming External Inter-Cell Handover Responses (Incoming External Inter-Cell
Handovers)
C2——Successful Incoming External Inter-Cell Handovers
A3——Outgoing External Inter-Cell Handover Requests
B3——Outgoing External Inter-Cell Handover Commands (Outgoing External Inter-Cell
Handovers)
C3——Successful Outgoing External Inter-Cell Handovers
Replaced with corresponding measurement points, the formulas for calculating different types of
HOSR can be as follows:
Success Rate of Handover: (C1<Successful Incoming Internal Inter-Cell Handovers> +C3)/
(A1<Incoming Internal Inter-Cell Handover Requests> +A3)
2012-12-12


第 10 页, 共 30 页

INTERNAL
Manual

Success Rate of Radio Handover: (C1 <Successful Incoming Internal Inter-Cell Handovers>
+C3)/(B1<Number of Incoming Internal Inter-Cell Handover Responses> +B3)
Success Rate of Intra-BSC Handover: C1/A1
Internal Radio Handover Success Ratio per cell: C1/B1
Success Rate of Incoming BSC Handover: C2/A2
Success Rate of Incoming BSC Radio Handover: C2/B2
Success Rate of Outgoing BSC Handover: C3/A3
Success Rate of Outgoing BSC Radio Handover: C3/B3

 Note:
If the BSC receives the Clear Command message sent by the MSC during an
inter-BSC handover, the current version does not count this case as a failed
handover. If a subscriber hangs up the phone during an intra-BSC handover, the
current version counts this case as a failed handover.

2 Influencing Factors
According to the cases and experience of the existing network, the factors that influence the
handover include the following types:

Hardware and transmission failures

Data configuration

Congestion

Coverage problems, and uplink and downlink imbalance

Interference

Clock problems

Failed inter-BSC/inter-MSC handovers
For details about all these factors, see section 3.2"Methods for Optimizing Handover Problems."

3 Analysis Process and Optimization Method
This chapter provides solutions to the problems about the handover when the following conditions
are all met:

The data configuration complies with the baseline of related parameters.

There is no problem about the engineering quality.

The coverage is good.

2012-12-12


第 11 页, 共 30 页

INTERNAL
Manual

3.1 Process of Analyzing Handover Problems
Generally, there are the following types of handover problems:

Call drops due to no timely occurrence of handovers

Failed handovers

Frequent (ping-pong) handovers

Poor downlink quality caused by slow handovers
These problems directly result in poor experience of end users, which is inclined to cause
complaints. Therefore, it is necessary to work out a method for optimizing the HOSR quickly or
even automatically to improve the network quality and user experience.

3.1.1 General Process of Locating a Handover Problem
3.1.1 shows the general process of locating a handover problem.

2012-12-12


第 12 页, 共 30 页

INTERNAL
Manual
The HOSR is low

Whether the
problem occurs in
some cells

No

Check whether the handover
thresholds and the BSC
clock over the A interface
circuit comply with the
standards

Yes
Whether a
hardware and transmission
failure occurs

Yes

Rectify the hardware and
transmission faults

Yes

Change the data
configuration

No
Whether the
data is configured
improperly
No
Whether the
target cell is congested

Yes

No
Whether a clock
problem occurs

Yes

Check whether a TRX or a
channel failure causes high
traffic in each timeslot. If yes,
change the TCH Traffic Busy
Threshold(%) and expand
the capacity

Clear the clock alarm and
replace relevant hardware

No
Whether an
interference problem
occurs

Yes

Optimize the RF

No
Whether a
coverage problem or uplink and
downlink imbalance occurs

Yes

Optimize the RF

No
Whether the
problem occurs in a co
-BSC
under a co-MSC

Yes

Check the common
hardware, the handover
data configuration between
the BSC and the MSC, and
the CGI of the cells

No
Perform automatic
neighboring cell
optimization

End

Figure 1.3 Flow chart of locating a handover problem

2012-12-12


第 13 页, 共 30 页

INTERNAL
Manual

3.2 Methods for Optimizing Handover Problems
Generally, handovers occur between two cells. The relationship between cells may be:

Between different BTSs in a BSC

In the same BTS in a BSC

Between different BSCs
Therefore, after you know how to locate and optimize the handover problems between two cells,
you can solve the handover problems in an entire network.
Handover problems may be caused by:

Hardware and transmission failures (bad TRXs or problems about the
combiner of the feeder and antenna system)

Improper data configuration

Congestion problems

Clock problems

Interference problems

Coverage problems, and uplink and downlink imbalance
If a low HOSR occurs, do as follows:
(1) Identify the problem.
(2) Troubleshoot the problem based on the factors such as hardware, data
configuration, congestion, clock, interference, and coverage
(3) Improve the HOSR according to the automatic optimization of neighboring
cells.

3.2.1 Classification of Handover Problems
I. Classification Description
Before analyzing the problem about the HOSR, determine the following points about handover
classification:
(1) Decide the scope of the failed handover. If the low HOSR occurs in all the
cells, check the problem from such aspects as the handover feature
parameters, the A interface circuits, and the BSC clock.
(2) If the low HOSR does not occur in all the cells, find out the TOP n poorest
cell. Then, proceed with the following steps specific to the cell.
(3) Distinguish whether there is any problem in the wireless interfaces
according to the differences between the HOSR and the Radio HOSR. The
Radio HOSR must be greater than or equal to the HOSR. If the HOSR is
much smaller than the Radio HOSR, analyze the problems about the
terrestrial links and the capacity. If the HOSR is a little different from the
Radio HOSR, consider the problems about the coverage and the
interference.
(4) Query the success rates of outgoing/incoming external/internal inter-cell
handovers in the handover performance measurement to analyze whether a
failed outgoing or incoming handover occurs. Analyze the performance
2012-12-12


第 14 页, 共 30 页

INTERNAL
Manual

measurement of outgoing and incoming external inter-cell handovers of the
faulty cell. From the performance measurement of outgoing external intercell handovers, find out to which cells the handover fails. Analyze counters
of the cells where incoming handovers fail, such as the Failed Incoming
External/Incoming Inter-Cell Handovers (No Channel Available), the Traffic
Volume on TCH, and the Congestion Ratio on TCH(All Channels Busy), to
decide whether the congestion of the target cell causes the failed handover.
(5) Query such counters as the TRX Availability and the TCH Availability of the
target cell to check whether any device is faulty.
(6) Query relevant alarms to analyze whether any terrestrial link device is faulty.

II. Traffic Measurement Analysis
By registering and analyzing the following counters, you can decide the scope and the basic cause
of a handover problem:
Cell Level
Incoming Internal Inter-Cell Handover Measurement
per Cell
Outgoing Internal Inter-Cell Handover Measurement
per Cell
Incoming External Inter-Cell Handover Measurement
per Cell
Outgoing External Inter-Cell Handover Measurement
per Cell
Incoming
Inter-RAT
Inter-Cell
Handover
Measurement per Cell
Outgoing
Inter-RAT
Inter-Cell
Handover
Measurement per Cell
Measurement of MRs upon Handover Initiation per
Cell
Channel Assignment Failure Measurement per Cell
Traffic Volume on TCH

3.2.2 Hardware and Transmission Failure
Symptom: The alarm system reports relevant alarm information. To rectify a hardware fault, clear
the alarms about the hardware failure. If the alarms are cleared, check the traffic measurement
information and analyze handover counters.
A hardware failure may involve the following hardware devices:

BTS transmission management unit

BTS TRXs

BTS combining and distribution unit

BTS feeder and antenna system
2012-12-12


第 15 页, 共 30 页

INTERNAL
Manual

I. Handling Process
(1) Check the data configuration of the hardware. If none of the data
configuration of the faulty cell and its neighboring cells is changed recently,
consider whether the handover problem is caused by a BTS hardware
failure.

If the handover problem occurs in only one cell under the BTS, consider
whether the problem is caused by the hardware failure of the cell. If a TRX is
damaged, a call fails to be handed over to this TRX.
If a similar problem also occurs in a co-site neighboring cell of this cell,
consider whether the problem is caused by the failure of the common hardware
of the cells, for example, the TMU failure.
You can block some TRXs to verify the preceding problems. If the HOSR
returns to normal after a TRX is blocked, check whether this TRX is faulty or
whether the CDU or the antenna related to this TRX is faulty.
If the uplink and downlink signals of a TRX are unbalanced, handover
problems such as frequent handover and lower HOSR often occur.
(2) Trace the Abis interface, and observe whether the signaling of the faulty cell
is normal and whether the uplink and downlink receiving quality in the
measure report is good. For detailed operations, refer to the M900&M1800
BSS Signaling Analysis Manual.

If the receiving level quality of half rate or full rate channel in the measurement
report is poor, the hardware of the cell is faulty or signaling cannot interact
normally due to serious interference in the cell. As a result, a handover problem
occurs.
Omitted.
III. Alarm Analysis
Observe whether any alarms with the following IDs are reported. If yes, refer to the BSS Alarm
Guide to handle the alarms.

Alarm ID and Name
4102
4104
4108
4114
4136
4144
4192

TRX LAPD Link Interrupt Alarm
TRX Config Mismatch Alarm
Radio link critical Alarm
TRX Interior I/O Alarm
TRX Hardware Critical Alarm
TRX VSWR alarm
TRX communication alarm
4714 E1/T1 Local Alarm
5286 CDU Level 1 VSWR Critical
Alarm
2012-12-12


第 16 页, 共 30 页

INTERNAL
Manual

5284 CDU Level 2 VSWR Critical
Alarm

5326 Level 1 VSWR alarm
5328 Level 2 VSWR alarm
3.2.3 Improper Data Configuration
I. Handling Process
Symptoms: An MS does not initiate any handover or frequently initiates handovers, which affects
the HOSR.
The handover parameters control the handover decision algorithm. If the handover parameters are
set improperly, the MS may not initiate any handover or frequently initiates handovers. In this
case, consider the cause from the following aspects:

Whether the PBGT HO Threshold in the data configuration is set properly
Avoid difficult handovers due to too great values of the handover thresholds or
frequent handovers due to the too small values. Proper settings can prevent
ping-pong handovers. For detailed settings of the thresholds, refer to the GSM
BSC6000 Performance Parameter Baseline (V900R008) (Chinese/English)
V2.0. Do not set the thresholds to the values deviating greatly from the baseline
values.

Whether the parameters related to the handover candidate cell in the data
configuration are set properly
Avoid the case that the MS cannot be handed over to a neighboring cell due to
the missed setting of the neighboring cell.

Whether the handover hysteresis parameters in the data configuration are
set properly
Avoid difficult handovers due to too large values of the handover hysteresis
parameters or frequent handovers due to too small values.

Whether the N and P counters in the data configuration are set properly
Avoid insensitive handover decision or difficult handovers due to the too large
values of the parameters, or the case that the target cell of a handover is not the
optimal due to the too small values of the parameters.
Avoid configuring the neighboring cells that share the same BCCH or the same
BSIC for a cell.
Abnormal circuit identification code (CIC) circuits may cause failed handovers.
For example, the CIC circuit allocated through a Handover REQ message received by the target
BSC is identified in the BLOCK state in the target BSC. Therefore, the BSC responds to the MSC
with a Handover Failure message whose cause value is Requested Terrestrial Resource
Unavailable. In this case, check the statuses of the circuits at the two sides of the A interface and
ensure that the circuits are in the same state.
You can trace the A interface signaling on the maintenance console to check whether the failed
handover is caused by the inconsistency of the circuit statuses. Do as follows:
(1) Trace the A interface signaling.
2012-12-12


第 17 页, 共 30 页

INTERNAL
Manual

(2) Filter the Handover Failure message.
(3) Check whether the cause value is Requested Terrestrial Resource
Unavailable.

II. Handover timer
When an abnormal handover occurs, promptly check the handover timer and ensure that the
handover timer is not less than the preset default value.
II lists the handover timers commonly used.
Table 3.1 List of handover timers commonly used
Timer
Name

Default Value
(ms)

Description

T7

10000

Timer for sending outgoing external inter-cell handover
requests and handover commands

T8

10000

Timer for running outgoing external inter-cell handover
commands and handover completion or clearance

T3103

10000

From the time when an intra-cell or inter-cell handover
command is executed to the time of an intra-cell or
inter-cell handover completion

T3105

70

During an asynchronous handover, from the time when
the BTS sends the MS a physical message to the time
when the BTS receives the set asynchronous balanced
mode (SABM) from the MS

T3124

320

During an asynchronous handover, from the time when
the MS sends the network a handover access Burst
message to the time when the MS receives a physical
message from the BTS

II shows the flow chart and the description of the timers.

2012-12-12


第 18 页, 共 30 页

INTERNAL
Manual

Figure 1.4 Flow chart and timer description

III. Traffic Measurement Analysis
Omitted.
IV. Alarm Analysis
Omitted.

3.2.4 Congestion of the Target Cell
I. Handling Process
Symptoms: After an MS initiates a handover request, the handover fails because no channel is
obtained.
The possible causes of cell congestion are:

The number of users in the cell soars and exceeds the designed number.

Improper settings of the network optimization parameters cause redundant
users in the cell.

Improper settings of the handover parameters cause the increase of the
users accessing the cell.
2012-12-12


第 19 页, 共 30 页

INTERNAL
Manual

After a handover fails because congestion occurs in the target cell, penalize the target cell to
prevent the MS from retrying to be handed over to this target cell. It is recommended that Penalty
Allowed be set to Yes.
Check whether the channel the congested cell is normal. If a TRX is faulty or a channel is
abnormal, rectify the relevant faults.
If full rate channels cannot be converted to half rate channels, it is recommended that you change
the channel attributes on the BSC6000 local maintenance terminal (LMT). That is, set the TCH
Rate Adjust Allow of all the TRXs under this cell to Yes. If the full rate channels can be
converted to half rate channels, properly reduce the value of TCH Traffic Busy Threshold(%) to
allocate half rate channels ahead of time and thus increase the system capacity. If the preceding
methods cannot solve the congestion problem, divide the cell or expand the capacity of the cell.
Since capacity expansion cannot be completed in a short time, you can set Channel Type to 1 or 2
to reserve channels for handovers. In this way, the failed handovers caused by congestion can be
reduced, and thus the HOSR improves.
(1) Register the measurement unit Channel Assignment Failure Measurement
per Cell. By analyzing the traffic measurement, you can be familiar with the
number of the times that all the channels are busy or that none of the
channels is configured when the BSC allocates SDCCHs, TCHFs, or
TCHHs in the processes such as immediate assignment, assignment,
internal intra-cell handover, incoming internal inter-cell handover, and
incoming external inter-cell handover.
(2) Change relevant parameters for the target cell according to the cause of the
failed handover.

If the failed handover is caused by the SDCCH congestion, set SDCCH Dynamic Allocation
Allowed to Yes.
If the failed handover is caused by the TCH congestion, reduce the value of the TCH Traffic
Busy Threshold(%) to allocate half rates ahead of time and thus relieve congestion. In addition,
you can set Channel Type to 1 or 2 to reserve channels for handovers.
SN

Measurement Counter

1

Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Immediate
Assignment Procedure (SDCCH)

2

Assignment Procedure (TCHF)

3

Assignment Procedure (TCHH)

4

Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Assignment
Procedure (TCHF/TCHH)

5

Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Internal IntraCell Handover Procedure (SDCCH)

2012-12-12


第 20 页, 共 30 页

INTERNAL
Manual

6

Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Internal IntraCell Handover Procedure (TCHF/TCHH)

7

Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Incoming
Internal Inter-Cell Handover Procedure (SDCCH)

8

Internal Inter-Cell Handover Procedure (TCHF/TCHH)

9

External Inter-Cell Handover Procedure (SDCCH)

10 Channel Assignment Failures (All Channels Busy or Channels Unconfigured) in Incoming
External Inter-Cell Handover Procedure (TCHF/TCHH)
11 Channel Assignment Failures (All Channels Busy or Channels Unconfigured) (SDCCH)
12 Channel Assignment Failures (All Channels Busy or Channels Unconfigured) (TCHF)
13 Channel Assignment Failures (All Channels Busy or Channels Unconfigured) (TCHH)
14 Channel Assignment Failures (All Channels Busy or Channels Unconfigured) (TCH)

III. Alarm Analysis
Omitted.

3.2.5 Clock Problems
I. Handling Process
Asynchronization and instability of the BTS clock are major causes of call drops during a
handover. Therefore, keep the BTS clock stable. Otherwise, handovers often fail and call drops
occur frequently.
A 13 MHz unlocked alarm is generated. The BSIC cannot be decoded. The HOSR of the
concerned cells decreases.
The clock source is abnormal and deviation may occur between the BTS clock and other BTS
clocks. As a result, MS abnormalities may occur during handovers.
To solve the problems about the unlocked clock and abnormality of the clock source, do as
follows:
(1) Check alarms. That is, check whether there is a 2214 E1 local alarm or
2216 E1 remote alarm. If there is, follow the concerned alarm handling
manual to handle the alarm. Then, observe the HOSR.
(2) Check the transmission link clock of the BTS. That is, use a frequency
meter to test the frequency deviation of the transmission link clock of the
BTS. If the frequency deviation is greater than or equal to 0.05 ppm, the
transmission link clock is abnormal, and the E1 or the optical transmission
link may be faulty or the clock source is faulty. Rectify the transmission link
fault through link-by-link self-loop until the alarm handling is complete.
(3) If the clock problem is not solved, reset the BTS (level-4) and observe
2012-12-12


第 21 页, 共 30 页

INTERNAL
Manual

alarms and the HOSR.
(4) If the problem remains unsolved, replace the TMU.

Omitted.
III. Alarm Analysis
Observe whether any alarms with the following IDs are reported. If yes, refer to the BSS Alarm
Guide to handle the alarms.

Alarm ID and Name
4154
4156
4184
4708
Alarm
4732
4734
4760

TRX main clock alarm
TRX slave clock alarm
TRX Clock Critical Alarm
Clock Reference Abnormal
TMU clock critical alarm
Master TMU clock alarm
13M Maintenance Alarm

3.2.6 Interference Problems
I. Handling Process
Severe interference in the network is inclined to cause the decrease in the receiving quality. As a
result, interference handovers or handovers in poor quality increase, the proportion of the power
budget (PBGT) decreases, and the quality of service (QoS) of the existing network is reduced to
some degree. Thus, user experience and the HOSR are affected.
Currently, the common interferences are intra-frequency and inter-frequency channel
interferences, Unicom CDMA interference, and mass multiplexing of the EGSM. If the idle Burst
function is not manually disabled after it is enabled, the interference of the entire network rises,
the noise floor increases, and the quality of the entire network decreases, thus affecting the HOSR.
The remote source signals of some optical fiber repeaters are inclined to cause intra-frequency
interference. Therefore, during optimization, you need to check the frequency of the source signals
and the frequencies of the cells close to the repeaters so that the frequency space is over 400 kHz.
If there is a repeater in the serving cell, do as follows:
(1) Choose Cell Software Parameters > Directly Magnifier Site Flag.
(2) Select Yes.
To solve interference problems, do as follows:
(1) Find out the cell or the frequency where large interference exists through
drive tests.
(2) Optimize the radio frequency (RF) by the following regular means:
2012-12-12

Adjust the tilt angle of the antenna.
Replace the frequency.

第 22 页, 共 30 页

INTERNAL
Manual

-

Change the transmit power and the coverage area of the cell.

You can also register the measurement results of interference fringes by auxiliary means to
estimate downlink interferences.
For more details about the solutions of interference problems, refer to the GSM Interference
Analysis Guide.
Omitted.
III. Alarm Analysis
Omitted.

3.2.7 Coverage Problems, and Uplink and Downlink Imbalance
I. Handling Process
Symptoms of signal coverage problems: The HOSR is low. Call drops occur frequently. There are
noises and metallic rings during conversations. The voice quality and the user experience are poor.
There are three types of signal coverage problems:

Low HOSR caused by cross coverage
Low values of the fringe thresholds, the large BTS power, and an improper tilt
angle cause cross coverage, thus forming intra-frequency interference and
affecting the HOSR.


Failed handovers caused by island effects

For example, the coverage area of the serving cell is much larger than that of its
neighboring cells, and the neighboring relation between the serving cell and the
neighboring cells of its neighboring cells is not configured. In this case, failed
handovers easily occur at the fringe of the serving cell.


Loopholes formed due to weak coverage

This section does not describe it in detail.
To solve signal coverage problems, do as follows:
(1) Find out the coverage problems in the existing network through drive test
reports of network optimization.
(2) Optimize the RF.
The low HOSR caused by uplink and downlink imbalance generally occurs when uplink signals
are weak. For example, there are problems in the hardware such as the CDU combiner, the uplink
channel loss is large, the uplink signals are weak, and the success rate of incoming external intercell handover is low. This low HOSR is generally caused by data problems (such as CGI errors in
the cell description data table, lack of measurement frequencies in BA list 1 and BA list 2, or intrafrequency and inter-frequency interferences), coverage dead zones in high traffic, or MS access
difficulties due to weak uplink signals. To test and analyze the low HOSR caused by uplink and
downlink imbalance, do as follows:
2012-12-12


第 23 页, 共 30 页

INTERNAL
Manual

(1) Check whether the hardware and maintenance boards of the relevant cell
are in the normal state, and whether there are any alarms about hardware
failures and the standing wave ratio (SWR). Refresh the channel status and
check whether the TCHs can be normally occupied.
(2) After hardware and channel problems are solved, check the handover data
configuration and ensure that the handover data complies with the
parameter baseline.
(3) Register the traffic measurement results of cell-level handovers. Check
whether the HOSR between some cells is always low.

1)

Make a field test for the cells where the HOSR is always low. That is,
perform a switchover and lock the main BCCH to act as the calling and
called parties respectively. Then decide the uplink and downlink problems
accordingly.

2)

If the uplink loss is large, it is recommended that you replace the combiner
to carry out an observation and a test.

Coverage problems and uplink and downlink imbalance are solved through RF optimization. For
detailed analysis, refer to the GSM BSS Network KPI (Coverage Problems) Optimization Manual
(V1.0).
Register the measurement unit Uplink-and-Downlink Balance Measurement per TRX about the
cells where the HOSR is low. Collect the uplink and downlink balance cases and carry out
analysis.
III. Alarm Analysis
Omitted.

3.2.8 Failed Inter-BSC/Inter-MSC Handovers
I. Handling Process
Symptoms: Inter-BSC or inter-MSC handovers fail.
The possible causes are that:

The data of the cells relevant to inter-MSC handovers is set wrongly.

The data of the cells relevant to the inter-BSC handovers is set wrongly.

The MSC and the BSC have different understandings of A interface
handover signaling. As a result, the cooperation at the A interface fails.

The clocks between BSCs are not synchronous.
To solve this problem, do as follows:
(1) Check whether the MSC data relevant to the cells where handovers fail is
set correctly, for example, the CGIs and the office direction of the cells. If
any data is set incorrectly, correct it and observe whether handovers
succeed.
2012-12-12


第 24 页, 共 30 页

INTERNAL
Manual

(2) Check whether the neighboring cells of the source and the destination BSC
are set correctly. If there is any abnormality, correct it and observe whether
handovers succeed.
(3) Trace A interface signaling. Check whether there is any abnormality in the
signaling cooperation of the handover process between the source BSC and
the MSC, and between the MSC and the destination BSC. For example,
check whether such a process that the MSC abnormally releases a
handover exists. If there is any abnormal process, find out the cause and
observe whether handovers succeed after such a problem is solved. For
detailed signaling analysis, refer to the M900&M1800 BSS Signaling
Analysis Manual.
(4) Check whether the source and the destination BSCs relevant to handovers
are locked with the clock of the upper-level MSC. If not, find out the cause
that the clock cannot be locked. Observe whether handovers succeed after
this problem is solved.

Omitted.
III. Alarm Analysis
Omitted.

3.2.9 Automatic Neighboring Cell Optimization
Currently, automatic neighboring cell optimization is the best method for optimizing the HOSR.
The method has been fully verified in the new functions of the MTN project. This optimization
method is to choose optimal neighboring cells for the serving cell through many times'
neighboring cell selection and tailor. This method can make the serving cell have neighboring cells
more close to the traffic model, thus avoiding failed handovers and call drops due to forced
configuration of neighboring cells according to geographical locations.
The premise of automatic neighboring cell optimization is to exclude objective factors such as
hardware problems, cross coverage, and uplink and downlink imbalance. After that, make it clear
to which neighboring cell the success rate of handovers from the serving cell is lower, and then
optimize this neighboring cell. The optimization method includes two types: parameter adjustment
and neighboring cell adjustment.
The process of automatic neighboring cell optimization is as follows:
(1) According to geographical locations, set as many neighboring cells as
possible for the serving cell. The upper limit is 32.
(2) Register the measurement unit GSM Cell to GSM Cell Outgoing Handover
Measurement, which period is 15 minutes.
(3) Observe the traffic measurement results. Exclude the cells that meet any of
the following conditions from the neighboring cells:

The HOSR is lower than 30%.
2012-12-12


第 25 页, 共 30 页

INTERNAL
Manual


The call drop rate is higher than 80%.



The handovers are relatively fewer according to the traffic, for example,
about 30 handovers every hour.

(4) After a neighboring cell is excluded, re-add a new neighboring cell
according to the timing advance (TA) principle from small to large, and
repeat the preceding steps.

4 shows the process of automatic neighboring optimization.

Perform optimization cyclically

Enter a neighboring cell list
(TA is limited)

Set the observation period and
the thresholds of the HOSR,
handover times, and call drops

Exclude irregular neighboring
cells

Update the list of neighboring
cells

Figure 1.5 Flow chart of automatic neighboring cell optimization

The TA is limited to six times as many as the average distance between sites. Do not consider the
cells with the TA beyond this threshold. You can flexibly set the lower limit of the HOSR and the
fewer limit of handover times.

3.2.10 Testing Tool Selection and Testing Suggestions
Generally, choose the industry-accepted and large-scale used TEMS as the testing tool. For the
cells where the HOSR is low, you need to carry out drive tests. In drive tests, the actual move
modes and habits of terminal users can be simulated. Therefore, drive tests play an important role
in neighboring cell optimization. Drive tests can avoid such risks as few handovers or low HOSR
caused by the addition of improper neighboring cells only according to geographical location
distribution on a map. Lay stress on and analyze any handover abnormality during a drive test.
Any of these abnormalities may be a cause of low HOSR.

3.2.11 Configuration Suggestions for Tests on the Existing Network
To configure the existing network according to different scenarios, refer to the GSM BSC6000
Performance Parameter Baseline (V900R008) (Chinese/English) V2.0. In the case of the low
HOSR, focus on checking the data configuration greatly different from the parameter baseline.

2012-12-12


第 26 页, 共 30 页

INTERNAL
Manual

4 Optimization Cases
4.1 A Handover Fails Because the BSIC Cannot Be
Decoded
During a drive test in an office, it is found that an MS cannot decode the BSIC of a neighboring
cell. As a result, the MS cannot initiate a handover even when the MS detects good levels of the
neighboring cell.
After analysis, it is learned that the PTCCH points to a wrong memory zone. As a result, some
MSs misunderstand that this channel is an FCCH, thus fail to synchronize the SCH and fail to
decode the BSIC. This product problem must be solved through version upgrade.

4.2 A Handover Fails Because Frequency Sequencing of
the MS Is Different from That of the BSC
In an office, field engineers check the TEMS drive test file and find that the cells of the main
BCCH at the EGSM cannot be handed over to the PGSM. It is checked that the rule of sequencing
frequencies at the MS side is different from that at the BSC side. When the serving cell is at the
EGSM and is configured with 1,800 neighboring cells, the MS sorts the EGSM first and then the
1,800 neighboring cells, whereas the BSC does in the contrary order. Thus, the different sequences
of neighboring cells at the BSC and MS sides cause a failed handover.
To mitigate such a problem, disable the Way of BA Delivery Optimized.

4.3 A Handover Fails Due to Unreasonable Parameter
Configuration
It is found in an office of BSC6000V9R8 that no bad quality (BQ) handover can occur. It is
checked that the Inter-cell HO Hysteresis of the serving cell is set to 63. The result of checking
codes shows that the level of the serving cell is equal to a value increased by 63 grades if Intercell HO Hysteresis is set to 63 when BQ HO Margin is set to the default value. Thus, the
calculated level of the serving cell is always larger than the level of any neighboring cell. As a
result, the handover cannot be initiated.
To solve such a problem, increase the value of BQ HO Margin to 127.

4.4 The Number of Failed Incoming BSC Handovers
Increases Because the Handover Request Does Not
Contain Class Mark 3
In a cell at the BSC boundary, the frequency of the main BCCH is configured as the PGSM and
that of the other TRXs is configured as the EGSM. Symptoms: The number of failed incoming
BSC handovers increases. The cause value is No Available Channel. The BSC6000 decides the
frequency supporting capability of an accessing MS according to class mark 3. If there is no class
2012-12-12


第 27 页, 共 30 页

INTERNAL
Manual

mark 3, the BSC considers that the MS supports only the frequency of the main BCCH. If the
Handover Request does not contain class mark 3 and the frequency of the other TRXs in the cell is
different from that of the main BCCH, accessing MSs are all allocated to the main BCCH. This
causes congestion and thus handovers fail. After the frequency of the other TRXs is changed to the
PGSM, the number of failed incoming BSC handovers caused by no available channel decreases
to 0. The problem is mitigated. According to the protocol, however, class mark 2 also has a field
for identifying whether an MS supports the EGSM or the RGSM (incapable of identifying the
DCS1800). Such a problem often occurs because some MSCs carry only class mark 2 in the
Handover Request or because an MS in a cell with the EGSM enabled does not report class mark
3.
To solve such a problem, upgrade the software version.

4.5 An Incoming BSC Handover Fails Because the A
Interface Phase Flag Is Set Wrongly
During an incoming BSC handover, after the BSC returns a Handover Request Ack message to the
MSC, the MSC returns a Clear Command immediately to clear call resources. The clearance cause
is Equipment Failure. The system responds to the MSC with a Handover Request Ack message
containing the speech version IE only when the A Interface Tag in the BSC attributes is
GSM_Phase_2+ and the BSC software parameter SpeechVer Send Flag In Ho Req Ack is set to
Yes. Since the A Interface Tag in the data configuration is GSM_Phase_2, the Handover Request
Ack message does not contain the speech version IE. Therefore, the MSC considers that this
message is illegal and sends a Clear Command.
To mitigate such a problem, set the A Interface Tag in the BSC attributes to GSM_Phase_2+ and
the BSC software parameter SpeechVer Send Flag In Ho Req Ack to Yes.

4.6 Because the Idle Burst Is Enabled, the Interference
Increases, the Receiving Quality Decreases, and the
HOSR Becomes Low
After an office is cut over, the drive test results show that the network quality decreases obviously
by about 3% to 4%. After it is confirmed that the problem is not caused by the problems about the
hardware, frequency planning, or cross engineering, it is found that the HOSR of the network is
2% to 3% lower than that the original network and other KPIs are normal.
After analysis, it is found that the idle burst function is enabled for testing. This function cannot
automatically be disabled. As a result, idle timeslots of TRXs are transmitted at full power, the
interference increases, the bit error ratio (BER) rises, and the receiving quality decreases.
Manually disable the idle burst function. The air interface quality of the entire network improves
effectively. The HOSR increases by 2% on the whole, almost the same as that of the original
network.

2012-12-12


第 28 页, 共 30 页

INTERNAL
Manual

4.7 Different HOSRs Resulting from Different Cause
Values Contained in the Clear Command Messages Sent
by Different Switches
There are multiple BSCs in office Z. Some BSCs are connected to Nortel switches and some are
connected to Ericsson switches. The whole HOSR of the BSCs connected to Ericsson switches is
2% to 3% lower than that of the BSCs connected to Nortel switches.
The result of the analysis on traffic measurement shows that the whole number of Failed Incoming
External Inter-Cell Handovers (Others) of the BSCs connected to Ericsson switches is very great
whereas the counter value of the BSCs connected to Nortel switches is 0. As defined in the
protocol, a Clear Command sent by the MSC can contain four cause values:

09: Call Control

0B: Handover Success

0A: Radio Interface Failure, Reverse to old Channel

01: Radio Interface Failure
For the failed handovers caused by MSC clearance, the Clear Command messages sent by
Ericsson switches contain the cause values of 0A and 01, whereas those sent by Nortel switches
contain the cause value of 09. Nortel switches do not send messages in accordance with the
protocol. Under Nortel switches, the failed handovers caused by MSC clearance are not collected
into the statistics. As a result, the HOSR of Nortel switches is higher.

5 Information Feedback
5.1 TEMS Log Files About Problem Cells
Cell information tables of TEMS tests are required to be provided in log files. The tables must be
in the format of *.cel.

5.2 Requirements of Configuration Data of the Existing
Network and Traffic Measurement Feedback
The latest data configuration and engineering parameter list are required. The feedback of traffic
measurement counters is required for consecutive two days. The following table lists the counter
types.
Call Drop Measurement per Cell
Incoming Internal Inter-Cell Handover Measurement per
Cell
Outgoing Internal Inter-Cell Handover Measurement per
Cell
Incoming External Inter-Cell Handover Measurement per
Cell
Outgoing External Inter-Cell Handover Measurement per
Cell
2012-12-12


第 29 页, 共 30 页

INTERNAL
Manual

Incoming Inter-RAT Inter-Cell Handover Measurement per
Cell
Outgoing Inter-RAT Inter-Cell Handover Measurement per
Cell
Measurement of MRs upon Handover Initiation per Cell
Channel Assignment Failure Measurement per Cell

2012-12-12


第 30 页, 共 30 页

Gsm bss-network-kpi-handover-success-rate-optimization-manual(HSR)

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to Gsm bss-network-kpi-handover-success-rate-optimization-manual(HSR) (10)

Recently uploaded (20)

Gsm bss-network-kpi-handover-success-rate-optimization-manual(HSR)