08 GSM BSS Network KPI (Immediate Assignment Success Rate) Optimization Manual

GSM BSS Network KPI (Immediate Assignment Success Rate) Optimization Manual INTERNAL

Product Name Confidentiality LevelGSM BSS INTERNALProduct Version Total 34 pages

GSM BSS Network KPI (Immediate

Assignment Success Rate) Optimization

Manual(For internal use only)

Prepared by GSM&UMTS Network Performance Research Department

Wang Fei

Date

2008-06-18

Reviewed by Date Reviewed by Date

Granted by Date

Huawei Technologies Co., LtdAll rights reserved

2008-10-25 All Rights Reserved Page 1 of 34


Revision Record

Date Revision Version

Change Description Author

2008-06-18 0.8 Draft completed. Wang Fei

2008-07-18 1.0 The document is modified according to review comments.

Wang Fei



References

Number References Author Date1 GSM BSS Network KPI (Immediate

Assignment Success Rate) BaselineLiu Xiuyu 2007-03-10



Content1 Overview of the Immediate Assignment Success Rate...........................................................71.1 Definition............................................................................................................................71.2 Signaling Procedure and Measurement Points.................................................................81.3 Symptom Description.........................................................................................................8

2 Factors Concerning Immediate Assignment Success Rate....................................................92.1 Signaling Factors...............................................................................................................92.2 Factor Analyzing According to the Definition of the KPI....................................................92.3 Factors Analysis...............................................................................................................10

2.3.1 Equipment Faults.........................................................................................................102.3.2 Um Interface Problems................................................................................................112.3.3 SDCCH Congestion.....................................................................................................112.3.4 MS Problems...............................................................................................................12

3 Method of Analyzing the Problem of Low Immediate Assignment Success Rate.................123.1 Process of Analyzing the Problem of Low Immediate Assignment Success Rate..........123.2 Method of Solving the Problem of Low Immediate Assignment Success Rate...............14

3.2.1 Equipment Faults.........................................................................................................143.2.2 Um Interface Problem.................................................................................................143.2.3 System Capacity (Congestion) Analysis.....................................................................193.2.4 MS Problem Analysis..................................................................................................24

4 Test Method...........................................................................................................................265 Cases of Immediate Assignment Success Rate Optimization..............................................275.1 Case 1: Decrease in Immediate Assignment Success Rate Because of SDCCH

Congestion Caused by Incorrect LAC Setting.......................................................................275.2 Case 2: SDCCH Congestion Caused by a Burst of Location Updates...........................285.3 Case 3: Low Assignment Success Rate Because the RACH Minimum Access Level Is

Set to 0...................................................................................................................................285.4 Case 4: Low Immediate Assignment Success Rate in Some Cells Because of MS Error

306 Onsite Information that Needs to Be Obtained......................................................................33



TablesTable 1 Mapping between formulae and factors..........................................................................10


GSGSM BSS Network KPI (Immediate Assignment Success Rate) Optimization Manual INTERNAL

GSM BSS Network KPI (Immediate Assignment

Success Rate) Optimization Manual

AbstractKey Words: SDCCH, immediate assignment success rate

Abstract: This document provides the procedure for analyzing the problems about immediate assignment success rate, describes the troubleshooting procedure, and provides typical cases.

Abbreviations:

Abbreviation Full SpellingBSC Base Station ControllerBTS Base Tranceiver Station SDCCH Stand-alone Dedicated Control

Channel

2023-4-8 Huawei Technologies Proprietary

Page 6 of 34


1 Overview of the Immediate Assignment Success Rate

Immediate assignment success rate indicates the success rate of the MS accessing the signaling channel. It concerns the procedure from the MS sending a channel required message to the BTS to the MS sending an establish indication message to the network.

Immediate assignment success rate is a key access counter. It directly reflects the

success rate of the MS accessing the signaling channel and affects the user experience.

The BSC versions concerned in this document are as follows:

BSC32——G3BSC32V300R007C01B015

BSC6000——BSC6000V900R001

1.1 Definition

The immediate assignment success rate is calculated from traffic statistics. The

recommended formula is as follows:

Immediate Assignment Success Rate = (Successful Immediate

Assignments/Immediate Assignment Requests) x 100%


Page 7 of 34


1.2 Signaling Procedure and Measurement Points

Note:

A1: Immediate Assignment Requests (Channel Requests (Circuit Service))

B1: Immediate Assignment Commands

C1: Successful Immediate Assignments (Call Setup Indications (Circuit Service))

1.3 Symptom Description

If the immediate assignment success rate decreases, the following symptoms may

occur:

Call setup success rate decreases.

Congestion occurs frequently on the SDCCH.


Page 8 of 34


Traffic volume on the TCH decreases.

Short messages cannot be successfully sent.

……

2 Factors Concerning Immediate Assignment Success Rate

2.1 Signaling Factors

According to the signaling flow, the factors concerning the immediate assignment

success rate during each phase are as follows:

(1) Random access request phase: whether the small interference signal on the Um

interface is regarded as the random access request by the BSC. Affecting factor:

Um interface problems;

(2) SDCCH assignment phase: whether the SDCCH is available after the random

access request is accepted. Affecting factor: SDCCH congestion;

(3) Channel activation phase: whether the channel activation is successful after the

channel assignment succeeds. Affecting factor: CHAN ACTIV NACK message is

received because of equipment faults during SDCCH activation;

(4) Immediate assignment command phase: whether the timer for the

acknowledgment of the immediate assignment expires. Affecting factor: Um

interface problems;

(5) Link establishment phase: whether the link can be established on the SDCCH

after the immediate assignment command is sent. Affecting factor: MS features.

The following analysis is based on these factors.


Page 9 of 34


2.2 Factor Analyzing According to the Definition of the KPI

As shown in the previous section, the following four factors are related to immediate

assignment success rate: Um interface problems, SDCCH congestion, reception of

CHAN ACTIV NACK because of equipment faults during SDCCH activation, MS

problems.

These factors are related to the definition of the KPI. There are three formulae that

can define the immediate assignment success rate, the mapping between formulae and

factors are as follows:

Table 1 Mapping between formulae and factors

Formula

Affecting Factor

Um interface problems

Reception of CHAN ACTIV NACK

because of

equipment faults during

SDCCH activation

SDCCH congestion

MS problems

Immediate Assignment Success Rate = (Successful Immediate Assignments/Immediate Assignment Requests) x 100%

√ √ √ √

Immediate Assignment Success Rate = (Successful Immediate Assignments/(Immediate Assignment Requests - (Immediate Assignment Commands - Successful Immediate Assignments))) x 100%

　 √ √ 　

Immediate Assignment Success Rate = (Successful Immediate Assignments/Immediate Assignment Commands) x 100%

√ 　　 √


Page 10 of 34


2.3 Factors Analysis

2.3.1 Equipment Faults

2.3.1.1 Low Immediate Assignment Success Rate Caused by TRX Faults

Generally, reception of CHAN ACTIV NACK during the SDCCH activation is caused

by equipment faults. If one TRX fails in the cell configured with multiple TRXs, the

immediate assignment may fail or the SDCCH may be congested. If you are sure that the

fault lies in a TRX, replace the faulty TRX. Otherwise, check the cables in the antenna

system and the VSWR. If the cables are properly connected and the VSWR is normal,

you need to replace the TRX, and then check whether services are normal again.

2.3.2 Um Interface Problems

2.3.2.1 Mistakenly Regarding Interference as Random Access Request

The BTS may mistakenly regard the interference on the Um interface as the random

access signal. This may lead to immediate assignment failure or SDCCH congestion. In

this situation, the interference problem should be solved.

2.3.2.2 Channel Immediate Assignment Failure Caused by Coverage Difference Between BCCH TRX and non-BCCH TRX

In a cell configured with six or more than six TRXs, the combination losses of the

BCCH TRX and non-BCCH TRXs are different, leading to coverage difference. If the

SDCCH is configured on the non-BCCH TRX, a call that is far away from the serving cell

may fail to access the SDCCH when it is assigned to the non-BCCH TRX. Thus, the call

drop may occur.

2.3.3 SDCCH Congestion

2.3.3.1 SDCCH Congestion Caused by Heavy Traffic

For SDCCH congestion caused by heavy traffic, capacity expansion is necessary.

You can also check the parameters concerning location update and SDCCH dynamic


Page 11 of 34


configuration. For SDCCH congestion caused by traffic burst, such as group sending of

short messages and location update at the portal of a tunnel, the problem cannot be

completely solved. You can, however, alleviate the congestion by enabling the functions

such as SDCCH dynamic adjustment.

2.3.3.2 Congestion Caused by Inappropriate Data Configuration

① Location area planning: a reasonable location area plan can help alleviate the

SDCCH congestion.

② Dynamic SDCCH assignment: the dynamic SDCCH assignment can help alleviate

SDCCH congestion.

③ Dual-band network: a reasonable setting of dual-band network parameters (for

example, CRO, CBA, CBQ, and Cell Reselection Hysteresis) can help alleviate SDCCH

congestion.

④ Check whether the settings of related timers (such as T3101, T3122, T3212, and

T3111) are reasonable.

2.3.4 MS Problems

In some cases, the location updates performed by the MS are abnormal, leading to a

low immediate assignment success rate. These MSs fail to establish a link on the

SDCCH after sending the channel request, causing a low immediate assignment success

rate. For the detailed analyzing method and case study, see case 4 in section 3.

3 Method of Analyzing the Problem of Low Immediate Assignment Success Rate

3.1 Process of Analyzing the Problem of Low Immediate Assignment Success Rate

When the immediate assignment success rate becomes low, you should firstly check


Page 12 of 34


the range in which the problem exists, and then find out the factors related to the problem

according to KPI statistical formulae. For details, see section 2.2 "Factor Analyzing

According to the Definition of the KPI". In the case of recommended formula, the general

process of analyzing the problem is as follows:


Page 13 of 34


Figure 1 Process of analyzing the problem of low immediate assignment success rate

3.2 Method of Solving the Problem of Low Immediate Assignment Success Rate

Before analyzing the cause of low immediate assignment success rate, you should

find out the difference between the actual immediate assignment success rate and the

expected value. You should also find out the influence of the problem and the definition

of the specific immediate assignment success rate KPI.

3.2.1 Equipment Faults

This part aims to solve TRX or transmission problems.

Firstly, you should check TRX Availability in BSC Measurement and SDCCH

Availability in SDCCH Measurement. Secondly, you should check the number of times

that a NACK message is received or the timer expires during the SDCCH activation. In

this way, you can determine whether the problem is caused by board faults.

You can also check whether hardware is faulty by viewing BTS alarms or by viewing the hardware state on Site Device Panel of the LMT. You can check the following traffic statistics for reference:

Cause BSC Level Cell Level

Equipment faults

BSC Measurement -> Access measurement per BSC ->SDCCH Availability per BSCConfigured SDCCHs per BSCAvailable SDCCHs per BSC

Call Measurement -> Channel Activation Measurement per Cell -> SDCCH Connection Measurement per Cell ->CHAN ACTIV NACK Messages Sent by BTS in Immediate Assignment Procedure (SDCCH)Channel Activation Timeouts in Immediate Assignment Procedure (SDCCH)KPI Measurement per Cell -> TCH Availability

3.2.2 Um Interface Problem

3.2.2.1 Mistakenly Regarding Interference as Random Access Requests

Interference may cause SDCCH congestion, causing low immediate assignment


Page 14 of 34


success rate. Especially in the areas with a small space between BTSs and dense BCCH frequency planning, the system allocates SDCCH for each signal if a large number of interference signals are received. Thus, the congestion may occur. In this situation, the immediate assignment success rate and paging success rate decrease, and the RACH may be overloaded. You can locate the problems by analyzing the idle interference bands and the TRX receive quality during the call.

If the system mistakenly regards the interference as a random access signal and

sends an assignment command, the immediate assignment will fail. You can locate the

problem by analyzing the Chan Req message.

For the SDCCH congestion caused by interference, you can check the following

traffic statistics for reference:

Cause BSC Level Cell Level TRX Level

SDCCH congestion caused by interference

BSC Measurement -> Access measurement per BSC ->Random Access Success Ratio per BSC

Call Measurement -> Flow Control Measurement per Cell -> MSG CCCH LOAD IND (RACH) Messages Sent on Abis Interface

MR Measurement ->Analyzed Measurement of Interference Band per TRXReceive Quality Measurement per TRX

For the problem of mistakenly regarding interference as random access

requests, you can locate the problem by performing signaling analysis on the Abis

interface of the cell with worst performance.

If the signaling analysis result shows that the random access signals are all from far

away, for example, the values of TA are higher than 10, sometimes even higher than

20, and the levels are lower than -100 dBm, the random access success rate is high

and the immediate assignment success rate is low, you can infer that the signals are

interference. To restrict the access of interference, reduce MAX TA and increase

RACH Min Access Level.

You can check the following traffic statistics for reference:

Cause BSC Level Cell LevelMistakenly Access of Interference

BSC Measurement -> Access measurement per BSC -> Random

Call Measurement -> Immediate Assignment Measurement per Cell -> Immediate Assignment Analyzed


Page 15 of 34


on the SDCCH

Access Success Ratio per BSC

Measurement per Cell -> Call Setup Indications Timed OutCall Measurement -> Immediate Assignment Measurement per Cell -> Immediate Assignment Analyzed Measurement per Cell -> Success Rate of Random Access

Figure 2 shows the contents of a channel required message.

Figure 2 Channel request signaling

Information element description:


Page 16 of 34


1. The meaning of Random Access Information (8bits) is as follows:

Table 9.9/3GPP TS 04.08: CHANNEL REQUEST message content

MS codes According to Establishment cause:bits8 .... 1

101xxxxx Emergency call

110xxxxx Call re-establishment; TCH/F was in use, orTCH/H was in use but the network does notset NECI bit to 1

011010xx Call re-establishment; TCH/H was in use and thenetwork sets NECI bit to 1

011011xx Call re-establishment; TCH/H + TCH/H was in useand the network sets NECI bit to 1

100xxxxx Answer to paging0010xxxx0011xxxx See table 9.9a/3GPP TS 04.080001xxxx

111xxxxx Originating call and TCH/F is needed, or originating calland the network does not set NECI bit to 1, orprocedures that can be completed with a SDCCH and thenetwork does not set NECI bit to 1. note 1

0100xxxx Originating speech call from dual-rate mobile station when TCH/H

is sufficient and supported by the MS for speech calls and the network sets NECI bit to 1 note 5

0101xxxx Originating data call from dual-rate mobile station when TCH/His sufficient and supported by the MS for data calls and the network sets NECI bit to 1 note 5

000xxxxx Location updating and the network does not set NECI bit to 1

0000xxxx Location updating and the network sets NECI bit to 1

0001xxxx Other procedures which can be completed with note 1an SDCCH and the network sets NECI bit to 1

011110xx One phase packet access with request for single timeslot uplink01111x0x transmission; one PDCH is needed.01111xx0


Page 17 of 34


01110xxx Single block packet access; one block period on a PDCH is needed for

two phase packet access or other RR signalling purpose.

01100111 LMU establishment note 2

01100xx0 Reserved for future use01100x0101100011 note 2a

01111111 Reserved for future use. note 2b

NOTE 1: Examples of these procedures are: IMSI detach, Short Message Service (SMS), Supplementary Service management, Location Services.

NOTE 2: If such messages are received by a network, an SDCCH shall be allocated.

NOTE 2a: If such messages are received by a network, an SDCCH may be allocated.

NOTE 2b: This value shall not be used by the mobile station on RACH. If such message is received by the network, it may be ignored.

2. The value of Access Delay indicates the TA.3. Bits 13, 14, 15, and 16 indicate the RACH level and the number of error bits in the training sequence. In this example, the value of bit 13 is 43. This indicates that the bits followed indicate the AB access level and the number of error bits in the training sequence. The value of bit 14 is 2. This indicates that the following two bits indicate the AB access level and the number of error bits in the training sequence. The value of bit 15 indicates the RACH level (39 – 120 = –81 dBm).

3.2.2.2 Coverage Difference

If the combination mode of the BCCH TRX and the non-BCCH TRX is different, or

the losses are inconsistent because of different transmit power or other problems, the

assignment on the non-BCCH TRX may fail.

The causes can be classified into three types:

① The transmit power of TRXs in the same cell is different.

If concentric cell technology is not applied, the power of different TRXs on the

antenna input port is different because of different uplink losses. This causes inter-TRX

coverage difference. Thus, assignment failure may occur. To solve this problem, check


Page 18 of 34


whether the combiner, divider, CDU, or SCU is properly connected.

② Multiple antennae are installed in a cell.

When multiple antennae are installed in a cell, the coverage differences of the

antennae may cause assignment failure. You can solve this problem through engineering

adjustment.

③ The transmit and receive antennae are not on the same horizontal plane or their

tilts are not the same. You can solve this problem by adjusting the antennae.

3.2.3 System Capacity (Congestion) Analysis

3.2.3.1 Congestion Caused by Heavy Traffic on the SDCCH

Generally, heavy traffic on the SDCCH is caused by frequent location updates. You

can check the rate of the SDCCH being occupied by querying the counter Channel

Requests (Location Updating) and the total number of channel requests. Figure 3

shows the ratios of different kinds of SDCCH requests to the total number of SDCCH

requests on Shanghai Unicom when the traffic is heavy. If the ratios are high, check the

location area planning and the drive test result. If the edges of location areas are

configured in the area with heavy traffic, reconfigure the location areas. In addition, you

can adjust the CHR parameters to reduce location updates. For details, see Case 1.

If the SDCCH congestion rate is high (greater than 5%), the traffic volume on each

SDCCH may have exceeded the limit as planned. You should check the configuration of

the location update and SDCCH dynamic configuration parameters. If the settings of

parameters are reasonable, a hardware capacity expansion is required.

If the congestion rate and traffic volume of the SDCCH are high, but the traffic

volume of the TCH is normal, the SDCCH congestion may be caused by burst traffic. For

the sites along the railway, especially for these near the tunnel portal, the capacity

configured is generally small. When a train passes through or stops, a large number of

location updates occur, which finally lead to SDCCH congestion. In addition, during the

period when the volume of short messages is large, SDCCH congestion may occur. This


Page 19 of 34


problem cannot be completely solved, but you can take some measures to alleviate the

congestion. For example, configure more SDCCHs, or enable the dynamic conversion

between the SDCCH and TCH. For details, see Case 2.

For the configuration of related parameters, see section 3.2.3.2 "Congestion Caused

by Inappropriate Parameter Configuration".

You can check the following traffic statistics for reference:

Cause BSC Level Cell Level

Congestion caused by heavy traffic on the SDCCH

BSC Measurement -> Access measurement per BSC -> Congestion Ratio on SDCCH per BSC

Call Measurement -> Immediate Assignment Measurement per Cell -> Channel Requests per Cell -> Channel Requests (MOC), Channel Requests (MTC), Channel Requests (Emergency Call), Channel Requests (Call Re-establishment), Channel Requests (Location Updating), Channel Requests (Packet Call), Channel Requests (LMU+Reserved), Channel Requests (Protocol Incompatible)KPI Measurement per Cell -> SDCCH Traffic Volume, SDCCH Seizure Requests, Failed SDCCH Seizures due to Busy SDCCH

Figure 3 shows the ratios of different kinds of SDCCH requests to the total number

of SDCCH requests on Shanghai Unicom when the traffic is heavy:

Rati os of Di ff erent Ki nds of SDCCH Requests to the Total Number of SDCCH Requests onShanghai Uni com when the Traffi c i s Heavy

12. 538%9. 309%

0. 055% 0. 030%

31. 059%

47. 003%

0. 001% 0. 006%0. 000%5. 000%

10. 000%15. 000%20. 000%25. 000%30. 000%35. 000%40. 000%45. 000%50. 000%

A300

ANu

mber

：of

SDC

CHre

ques

ts(m

obil

eor

igin

ated

call

)A3

00C

Numb

er：

of S

DCCH

requ

ests

(mob

ile

term

inat

edca

ll)

A300

DNu

mber

：of

SDC

CHre

ques

ts(e

merg

ency

call

)

A300

ENu

mber

：of

SDC

CHre

ques

ts (

call

re-

esta

blis

hmen

t)

A300

FNu

mber

：of

SDC

CHre

ques

ts(l

ocat

ion

upda

ting

)

A300

HNu

mber

：of

SDC

CHre

ques

ts(p

acke

t ca

ll)

A300

INu

mber

：of

SDC

CHre

ques

ts (

LMU

+Re

serv

ed)

A300

KNu

mber

：of

SDC

CHre

ques

ts(p

roto

col

inco

mpat

ible

)Rati os of di ff erent ki nds of SDCCH requests

Figure 3 Ratios of different kinds of SDCCH requests to the total number of SDCCH requests on Shanghai Unicom when the traffic is heavy


Page 20 of 34


Congestion Caused by Inappropriate Parameter Configuration

An inappropriate parameter configuration may lead to the SDCCH congestion.

The parameters related to SDCCH congestion and their configurations are as

follows:

1. CRO: Generally, it is recommended that an MS should not camp on a cell with

high traffic volume or low QoS. The parameters of such cells should be set as follows:

firstly, set PT to 31, so that TO becomes invalid. Secondly, set C2 to the same value as

CRO subtracted from C1 so that the value of C2 is reduced and thus the possibility of

reselecting this cell is reduced. In addition, network operators can set the CRO according to

the actual conditions. The greater the CRO is, the more difficult it is for an MS to access

the cell.

2. Cell Reselection Hysteresis (CRH): This parameter is used to determine

whether cell reselection is performed between different location areas. This parameter

can prevent frequent location update, thus minimize the possibility of losing paging

messages. Generally, this parameter is set to 6 dB. For dual-band networks in urban

areas not sharing the location areas, this parameter is set to 8–10 dB.

If the traffic volume in an area is high, and signaling overload occurs frequently,

it is recommend that you set the CRH of the neighboring cells belong to different

LACs under this area to a larger value.

If the overlapping area of the neighbor cells under different location areas is

large, it is recommended that you set the CRH of the cells to a larger value.

If some areas between two neighbor cells under different location areas have

poor coverage, or if most of the MSs in the area (such as highways) move at a

high speed, it is recommended that you set the CRH of the cells to 2–6 dB.

3. Set the parameters related to SDCCH dynamic adjustment as follows:

Parameter DescriptionSDCCH Dynamic

If this parameter is set to YES, the dynamic conversion between the SDCCH and the TCH is enabled.


Page 21 of 34


Allocation Allowed

Idle SDCCH Threshold N1

According to the channel assignment algorithm, when the number of idle SDCCHs in the cell is smaller than or equal to the value of this parameter, the system tries to find a TCHF that can be converted into an SDCCH. This parameter is one of the conditions for the dynamic conversion from TCHF to SDCCH.

Cell SDCCH Channel Maximum

According to the channel assignment algorithm, the system determines whether the number of SDCCHs in the cell after the conversion exceeds the value of Cell SDCCH Channel Maximum before initiating the dynamic conversion from TCH to SDCCH. If yes, the conversion is not initiated.

TCH Minimum Recovery Time(s)

This parameter specifies the minimum time for the recovery of a TCH from the SDCCH.

4. Related timers and their recommended configurations are as follows:

T3101: This timer is used for immediate assignment procedure monitoring.

Reduce the value of this parameter properly to minimize the congestion caused

by dual assignment of the SDCCH. If the length of this timer is too long, the

invalid usage of signaling resources is increased. To enhance the usage of the

signaling resources, you should shorten the length of this timer, especially when

the queuing function is enabled.

T3122: This timer is started by the MS when the MS receives the IMMEDIATE

ASSIGN REJECT message. The MS can send a new channel request message

only after T3122 expires. If the length of T3122 is too short, the MS may send

the channel request message frequently when no system resources are

available. This increases the load on the RACH and CCCH.

T3212: This timer is used for the periodic location update. Increase the length of

this timer properly to reduce the load on the SDCCH brought by periodic

location update.


Page 22 of 34


T3111: This timer delays the deactivation of channels after main signaling links

are disconnected to reserve time for repeated link disconnection. T3111 is

started during the disconnection of both TCH and SDCCH. The value of T3111

must be the same as the value of T3110 at the MS side. Generally, they are set

to 2 seconds. If the length of T3111 is too long, the SDCCH congestion rate is

increased.

5. RACH Min Access Level: If the value of this parameter is too small, much

interference may access the network, thus leading to SDCCH congestion. If the value of

this parameter is too large, it is probable that MSs cannot make the call even though the

signal level is high. This parameter should be set on the basis of the actual receiver

sensitivity of the BTS, the minimum access level of the MS, and the interference

condition.

6. Late assignment: This function is set at the MSC side. If this function is enabled,

the call is on the SDCCH before the called MS answers. This prolongs the time when the

SDCCH is occupied and thus may cause SDCCH congestion.

7. RXLEV_ACCESS_MIN: If this level is too low, a large number of MSs may access

the network, and SDCCH congestion may occur during call initiation or location update.

In addition, interference signal may be mistakenly regarded as the random access

signals. For details, see Case 3.

8. Tx-integer: When the network traffic is heavy, the success rate of immediate

assignment is low if the sum of S and T is small. Thus, the value of T should be properly

adjusted to make the sum of S and T greater. For details about the value of S, see the

description of Tx-integer.

9. Cell Bar Access (CBA):

0: cell access allowed;

1: cell access prohibited.

Together with CBQ, CBA determines the priority of a cell. For details, see the GSM

04.08 protocol.


Page 23 of 34


CBQ CBACell Selection Priority

Cell Reselection Priority

0 0 Normal Normal0 1 Prohibited Prohibited1 0 Low Normal1 1 Low Normal

10. Cell Bar Qualify (CBQ): CBQ takes effect during cell selection rather than cell reselection.

3.2.4 MS Problem Analysis

3.2.4.1 Checking the Problem Symptoms

Objective: to check the problems that exist on site and solve the problems caused by

poor air interface quality and incorrect parameter settings.

The symptoms are listed as follows:

Symptom 1: The number of failed SDCCH seizures due to timeout is almost equal to

the number of location updating requests minus the number of link establishment

indications for location updating.

Symptom 2: The problem happens discontinuously during both busy hours and idle hours.

Symptom 3: The call service is normal. Except for low immediate assignment

success rate in a few cells during certain periods, the KPIs and drive test result are

normal, and no call drop complaint is received. Because when an MS fails in location

update, it may try to access other cells or try to access the cell again periodically.

Symptom 4: No interference and no cells with the same BCCH frequency and BSIC

exist.

Symptom 5: No abnormality such as uplink-downlink imbalance exists.

Symptom 6: The result of signaling analysis on the Abis interface shows that the

number of request retransmissions and the retransmission interval of the failed location

updates meet the configuration requirements.

If all the previous symptoms exist, you can infer that the problem is low immediate

assignment success rate caused by abnormal location update.


Page 24 of 34


3.2.4.2 Problem Handling

As the problem of low immediate assignment success rate is caused by some MSs,

and the MSs cannot be located because no layer-3 information is provided in location

update procedure, this problem cannot be solved on the network side currently.

To improve the user satisfaction, do as follows:

(1) Reduce the impact of abnormal MS location updates on the network KPIs.

(2) Communicate thoroughly with customers.

Problem handling:

If any of the previous six symptoms does not exist, initiate the procedures for

network optimization and troubleshooting.

To perform network optimization according to the RSL signaling, do as follows:1. If the value of TA in the location update signaling is high and the access level is

low during the time when the problem exists, do as follows:

Adjust the following parameters: Max TA, RACH Min Access Level, and Tx-

integer (increase S and T).

Increase the length of the following timers: T3101, T3122, T3212, and T3111.

Adjust the cell selection priority by setting the value of CBA and CBQ.

2. For the sites located on the edge of location areas, do as follows to optimize the

location update strategy:

Adjust the parameter Cell Reselection Hysteresis and adjust the cell selection

priority by setting the value of CBA and CBQ.

3. Reduce the MS retransmission times (recommended value: 1) and reduce the

impact of retransmission on the network KPIs.

Data analysis:

1. For the analysis report of problem location, see the following attachment.

Based on the previous analysis, we find that the channel requests received when the

problem occurred were sent by some abnormal MSs, and the BTS responded and

handled the requests correctly. The analysis is based on the uplink and downlink. The


Page 25 of 34


details are described as follows:

2. Uplink – channel requests from the MS

(1) When the cell access is prohibited, the immediate assignment success rate of the

cell becomes normal. This indicates that the abnormal location updates are caused by

the MS rather than the interference.

(2) The location update requests are with high level, low error rate, and small TA

(except for some sites with wide coverage area). This indicates that the abnormal

location updates are initiated by the MS.

(3) The number of location update requests sent by an MS complies with the settings

of network parameters. This also indicates that the abnormal location updates are

initiated by the MS.

3. Downlink – immediate assignment

(1) The immediate assignment commands are sent correctly, because the values of

T1, T2, and T3 are the same as those in the channel requests.

(2) Call access and the supplementary services such as short message are all

normal. The problem only exists in the SDCCH access related to location update. This

indicates that the immediate assignment commands are sent correctly.

4. Network data comparison (for the swapped offices and the offices with segmental

networking)

By comparing the data before and after swapping (for swapped offices) and the data

of the surrounding cells (for the offices with segmental networking), we find that the

problem also exists in the networks of other companies.

4 Test Method

Immediate assignment success rate is a traffic measurement counter.

As the formulae of this counter in different manufacturers and operators are different

currently, you should record the value of other related counters and choose a proper


Page 26 of 34


formula according to the actual conditions.

5 Cases of Immediate Assignment Success Rate Optimization

5.1 Case 1: Decrease in Immediate Assignment Success Rate Because of SDCCH Congestion Caused by Incorrect LAC Setting

Problem description

The SDCCH congestion rate of cell 2 in S1/1/1 configuration was higher than 8%,

and the immediate assignment success rate of that cell was 90%.

Cause analysis and handling

(1) The measurement counters of the TCH and SDCCH shows that the traffic volume on

the TCH is low, and the traffic volume per cell in busy hours is lower than 2.2 Erl;

however, there are many SDCCH seizure requests (3032 in busy hours), the traffic

volume on the SDCCH is 1.86 Erl, and the congestion rate is higher than 8%.

(2) The SDCCH is mainly occupied by the signaling before the call is established,

the signaling during handover, the signaling for location update in idle mode, and the

short messages.

(3) As the traffic volume on the TCH, the number of TCH seizure requests (318), and

the number of handover requests (146) are all normal, we infer that the high SDCCH

seizure rate may be caused by too frequent location updates or too many short

messages.

(4) The LAC of the cell is 0500, but the LACs of other cells around this cell are all

0520. After the LAC of the cell is changed to 0520, the number of SDCCH seizure

requests becomes 298, the traffic volume on the SDCCH becomes 0.27 Erl, the

congestion rate decreases to 0, and the immediate assignment success rate reaches up

to 95%.


Page 27 of 34


5.2 Case 2: SDCCH Congestion Caused by a Burst of Location Updates

Problem description

The immediate assignment success rate was low in a local network. Traffic statistics

showed that the problem was mainly caused by SDCCH congestion on some sites.


(1) Traffic statistics shows that the SDCCH is occupied for 300 to 400 times in busy

hours. The congested cells are all in S1/1/1 configuration and each cell is configured with

eight SDCCHs/8. In normal conditions, 300 to 400 times of SDCCH seizures are

acceptable, but SDCCH congestion happens for tens of times in each cell.

(2) Traffic statistics show that most of the SDCCH seizures are caused by location

update. As most of the congested BTSs are located in the intersection areas of two

location areas along the railway, we doubt that the SDCCH congestion may be caused

by a burst of location updates.

(3) After checking the five-minute traffic statistics, we find that most of the location

updates happened within a certain period of five minutes. The train timetable shows that

four or five trains had passed within that five minutes. When the trains passed, a burst of

location updates were performed during a short period of time, thus, the congestion

happened.

(4) For the BTSs located in the intersection areas of two location areas along the

railway, it is recommended that the SDCCH dynamic assignment function be enabled

and adequate SDCCHs be configured.

5.3 Case 3: Low Assignment Success Rate Because the RACH Minimum Access Level Is Set to 0

Problem description

Subscribers of the cells controlled by a certain BSC complained that the success

rate of random access was low.

Cause analysis and handling:


Page 28 of 34


(1) The clock status of the BTSs are normal.

(2) The value of parameter Tx-integer (RACH Timeslot) is normal. By analyzing

the data and traffic statistics, we find that the parameter RACH Min Access Level

of all the cells with low assignment success rate is set to 0. The immediate

assignment success rates of cells whose RACH Min Access Level are set to 1, 3,

or 5, however, are around 99%. According to the previous data, we infer that the

problem is caused by mistaken access.

(3) If RACH Min Access Level is set to 0, there is no restriction on the random

access; If RACH Min Access Level is set to 1, a call can access the cell only

when the level is higher than –109 dBm.

In Shantou Unicom, the RACH Min Access Level of most of the cells is set to 0 to

improve the paging success rate. This causes a great number of mistaken access

and finally causes low assignment access rate.

In addition, the parameter Random Access Error Threshold also has error

restrictions on the access signals.

(4) When RACH Min Access Level and Random Access Error Threshold of all

the cells in module 1 of the BSC in Shantou Unicom are set to different values, the

results are as follows:

When Random Access Error Threshold and RACH Min Access Level are

set to 200 and 1 respectively, the mean immediate assignment success rate

of all the cells is 97.9% an hour later;



of all the cells is 91.92% an hour later;



of all the cells is 97.24% three hours later.

After discussing with the customer, we decide to set Random Access Error


Page 29 of 34


Threshold and RACH Min Access Level of all the cells to 180 and 3

respectively. In this way, mean immediate assignment success rate of all the

cells is higher than 98.6%.

5.4 Case 4: Low Immediate Assignment Success Rate in Some Cells Because of MS Error

Problem description

The personnel on site reported that the immediate assignment success rate of some

cells was low.


(1) As the problem exists only in some cells, it may be caused by the MS error.

(2) Traffic statistics of the cells with the problem show that the number of link

establishing failures during location updates is approximately equal to the number of

times that the setup indication timer expires, as shown in the following figure. This

indicates that the SDCCH setup failure is mainly caused by location updates.

(3) The signaling of the cells with problem shows that generally the value of TA is


Page 30 of 34


small and the level is high. This indicates that the problem is not caused by interference.

But after the BSS sends the immediate assignment command and waits for the MS to

access, the uplink signal strength detected by the BTS is always lower than –110 dBm.

This indicates that the MS does not report the link establishing indication and thus

causing the immediate assignment failure.

In addition, we check whether location updates are initiated by a same MS and

whether the number of location updates is in compliance with the network setting. The

following figure shows the signaling captured in the local network. As can be seen, the

location updates are mainly initiated by a same MS and the maximum number of

retransmissions of the location update request set on the network is five.


Page 31 of 34


The parameters associated to channel request are configured as follows:

(4) Cell bar access (CBA) status modification test.

After the CBA of a cell is enabled, the system information of the cell will carry a flag

bit. The MSs that receive such system information will not consider that cell during cell

reselection, and the number of location update requests to the neighboring cells

increases.


Page 32 of 34


After the CBA is enabled, the immediate assignment KPIs become better, as shown

in the following figure.

(5) For a swapping project, check whether the problem exists in the network before

swapping. If the problem existed, then we can infer that the problem is mainly

related to the MS.

6 Onsite Information that Needs to Be Obtained

1. Traffic statistics of the network during one week before and after the problem

occurred.

BSC32:

(1) All the counters of the Access Capability Measurement 1 and Access

Capability Measurement 2 in BSC Overall Performance Measurement;

(2) All the counters of Receive Quality Performance Measurement;

(3) All the counters of the SDCCH Assignment Performance Measurement

and TCH Assignment Performance Measurement in Channel Assignment

Performance Measurement;

(4) All the counters of SDCCH Performance Measurement, TCH

Performance Measurement, and Random Access Performance


Page 33 of 34


Measurement in Cell Performance Measurement.

BSC6000:

(1) All the counters of Access Measurement per BSC in BSC Measurement;

(2) All the counters of Interference Band Measurement per TRX and Receive

Quality Measurement per TRX in MR Measurement;

(3) All the counters of Channel Configuration Measurement per Cell in

Channel Measurement;

(4) All the counters of KPI Measurement per Cell, Immediate Assignment

Measurement per Cell, and Flow Control Measurement per Cell in Call

Measurement.

2. Data configuration and OMC data operation logs before and after the problem

occurred.

3. Data configuration file and engineering parameters table.

4. Whether the network optimization operations, such as antenna adjustment,

frequency adjustment, and location area adjustment, are performed after the problem

occurred.

5. Alarms generated before and after the problem occurred, including commissioning

alarms.


Page 34 of 34

08 GSM BSS Network KPI (Immediate Assignment Success Rate) Optimization Manual

Documents

erent ki nds

huawei technologies

chan activ

assignment

drive test

rate mobile

channel assignment

flow control