Gsm Handover Problem Analysis

P-1

Course Name

Case Analysis —— Handover www.huawei.com

Copyright © 2010 Huawei Technologies Co., Ltd. All rights reserved.

Confidential Information of Huawei. No Spreading Without Permission

P-2

Course Name

References z

BSC Counter Reference(V900R008C12)

z

BSS KPI Reference(V900R008C12)


Page2


P-3

Course Name

Objectives z

Upon completion of this course, you will be able to:

Understand the Measurement Points and Performance Index of handover

Analyze, locate and solve handover problems


Page3


P-4

Course Name

Contents 1. Handover Process Flow 2. Measurement Points and Performance Index of Handover 3. Analysis of Handover Problem 4. Handover Cases


Page4


P-5

Course Name

Handover Process Flow MSC2 Search target cell（7）

Handover judgement（4）

BSC1

MSC1 Inter-MSC H.O （8）

CGI（6）

H.O request（7）

BTS1

BA2 table（1）

MS

Channel active （5） Preprocessed MR （3）

z

z

z

z z

z

z

BTS2

MR（2）


z

BSC2

Page5

BCCH frequencies of all adjacent cells in BA2 table are sent to MS on system message. MS reports measurement report to BSC. It includes the BCCH , BSIC and signal level of the adjacent cells and serving cells. When the measurement report is preprocessed, BSC identifies the CGI of all adjacent cells through BCCH frequency and BSIC . BSC executes handover judgment flow such as basic cell ranking. Once a proper target cell is found, the handover request message which includes the target cell CGI will be sent to BSC. If the target cell is an internal cell, BSC send the “channel active” to the BTS. If the target cell is an external cell, both the CGI of the target cell and that of service cell are sent to MSC via the handover required. By matching the CGI of the target cell, MSC searches the target cell. Once the cell is found, MSC will confirm which BSC is belonged to, and send the handover request message to this BSC. If there is no CGI of the target cell in local MSC, MSC will check “Adjacent MSC Table" and search the target MSC, then send the handover request message to that MSC.


P-6

Course Name



Page6


P-7

Course Name

Measurement Points of Intra BSC Handover MS

BTS1(target)

BSC

BTS2 or BTS1(original)

MS

MSC

Measure Report from MS

Channel_Acive

CH300: Internal Intra-Cell Handover Requests CH320: Incoming Internal Inter-Cell Handover Requests CH310: Outgoing Internal Inter-Cell Handover Requests

Channel_Active_Ack Handover Command (Old FACCH) Handover Access (New FACCH) Handover_Detect PHYINFO PHYINFO First SABM UA

CH301: Internal Intra-Cell Handover Commands CH321: Incoming Internal Inter-Cell Handover Responses (Incoming Internal Inter-Cell Handovers) CH311: Outgoing Internal Inter-Cell Handover Commands

Establish_IND

Handover Complete (New FACCH)

CH303: Successful Internal Intra-Cell Handovers CH323: Successful Incoming Internal Inter-Cell Handovers CH313:Successful Outgoing Internal Inter-Cell Handover

Handover_Performed


Page7


P-8

Course Name

Performance Index of Intra BSC Handover (Cell) RH303B:Internal Handover Success Ratio per Cell ＝ CH303: Successful Internal CH323: Successful Incoming Internal Inter-Cell Intra-Cell Handovers + Handovers CH300: Internal Intra-Cell Handover Requests

+

CH320:Incoming Internal Inter-Cell Handover Requests

×100％

Internal Handover Success Ratio per Cell <= IntraIntra-BSC Radio Handover Success Rate RH303C: Intra-BSC Radio Handover Success Rate ＝ CH303: Successful Internal CH323: Successful Incoming Internal Inter-Cell Intra-Cell Handovers + Handovers CH321:Incoming Internal CH301: Internal Intra-Cell Inter-Cell Handover Handover Commands+ Responses


z

z

z z

z

×100％

Page8

Difference between " Internal Handover Success Ratio per Cell " and " Intra-BSC Radio Handover Success Rate " : As viewed from formulas, both numerators are success times of handover, while the denominators are different. While viewed from the measurement points of the counter, intra-cell/inter-cell handovers requests >= intra-cell/inter-cell cell handover commands or responses, so Internal Handover Success Ratio per Cell <= Intra-BSC Radio Handover Success Rate. In practice: If "Internal Handover Success Ratio per Cell“ = "Intra-BSC Radio Handover Success Rate", this indicates that there is no problem with data, the unsuccessful handovers are caused by radio interfaces, and the radio reasons like interference and coverage should be checked in the first place. If "Internal Handover Success Ratio per Cell“ < "Intra-BSC Radio Handover Success Rate", this indicates that failures may exist in the process from “Channel_Activate” to "HOCOMD", and there might be data problem or congestion.


P-9

Course Name

Performance Index of Intra BSC Handover (BSC) ZK3174: Internal Handover Success Ratio per BSC ＝ ZCH303:Successful Internal Intra-Cell Handovers per BSC

ZCH323: Successful Internal Incoming Cell Handovers per BSC ×100％ ZCH300: Internal Intra-Cell ZCH320: Internal Incoming Cell Handover Requests + Handover Requests per per BSC BSC

+

Internal Handover Success Ratio per BSC <= Radio Handover Success Ratio per BSC ZK3175: Radio Handover Success Ratio per BSC＝ ZCH303:Successful Internal Intra-Cell Handovers per BSC

ZCH323: Successful Internal Incoming Cell Handovers per BSC

+

ZCH301:Internal Intra-Cell Handover Commands per BSC

+

ZK3190:Internal Incoming Cell Handover Commands per BSC

×100％

N

Internal Incoming Cell Handover Commands

per BSC ＝ ∑ Incoming Internal Inter - Cell Handover Responses i=0


Page9


P-10

Course Name

Measurement Points of Inter-BSC Handover MS

BTS(Original)

BSC(Original) MSC

BSC(Target)

BTS(Target)

MS

Measurement Report Measurement Report Handover Required CH330: Outgoing External Handover Request Inter-Cell Handover Requests

CH340:Incoming External Inter-Cell Handover Requests

Channel_Active Channel_Active_ACK Handover_Request_ACK Handover Command

CH341:Incoming External Inter-Cell Handover Responses

CH331:Outgoing External InterCell Handover Commands

Handover Access Handover Detect Handover Complete Handover Complete CH343:Successful Incoming External Inter-Cell Handovers

CH333:Successful Outgoing External Inter-Cell Handovers

Clear Command (HO Successful)

RF Channel Release Clear Complete


z

z

Page10

The key measurement points are as follows, same as those of intra BSC handover: 1. After sending "HO-REQUIRED", the original BSC measures "Attempted outgoing inter BSC inter cell handovers". 2. After receiving "HO-REQUEST", the target BSC measures "Attempted incoming inter BSC handovers". 3. After sending "HO-REQUEST ACK", the target BSC measures "incoming inter BSC handovers". 4. After receiving "HO-Command" , the original BSC measures "outgoing inter BSC handovers". 5. After receiving "HO-Complete", the target BSC measures "Successful incoming inter BSC handover" 6. After receiving "Clear-COM" and the cause value is "HO-Successful", the original BSC measures "Successful outgoing inter BSC inter cell handover". The difference between "handover times" and "handover request times": Handover times - After "HO-COM" is received or "HO-REQ-ACK" is sent Handover request times - After "HO-REQUIRED" is sent or "HO-REQUEST" is received Therefore, inter BSC inter cell radio handover success rate >= inter BSC inter cell handover success rate.


P-11

Course Name

Performance Index of Incoming External Inter-Cell Handover(Cell) TH343: Success Rate of Incoming External Inter-Cell Handovers＝ CH343:Successful Incoming External Inter-Cell Handovers CH340:Incoming External Inter-Cell Handover Requests

×100％

Success Rate of Incoming External InterInter-Cell Handovers <= External Incoming Radio Handover Success Ratio per cell RH303D: External Incoming Radio Handover Success Ratio per cell ＝ CH343:Successful Incoming External Inter-Cell Handovers CH341:Incoming External Inter-Cell Handover Responses


×100％

Page11


P-12

Course Name

Performance Index of Incoming External Inter-Cell Handover(BSC) ZK3178: External Incoming Cell Handover Success Ratio per BSC ＝ ZCH343:Successful External Incoming Cell Handovers per BSC ZCH340:External Incoming Cell Handover Requests per BSC

×100％

External Incoming Cell Handover Success Ratio per BSC <= External Incoming Cell Radio Handover Success Ratio per BSC ZK3179: External Incoming Cell Radio Handover Success Ratio per BSC＝ ZCH343:Successful External Incoming Cell Handovers per BSC ZK3191:External Incoming Cell Handover Commands per BSC

×100％

N

External Incoming Cell Handover Commands per BSC ＝∑ Incoming External Inter - Cell Handover Responses i =0


Page12


P-13

Course Name

Performance Index of Outgoing External Inter-Cell Handover(Cell) TH333: Success Rate of Outgoing External Inter-Cell Handover ＝

CH333: Successful Outgoing External Inter-Cell Handovers CH330:Outgoing External Inter-Cell Handover Requests

×100％

Success Rate of Outgoing External InterInter-Cell Handover <= External Outgoing Radio Handover Success Ratio per cell RH303E: External Outgoing Radio Handover Success Ratio per cell＝

CH333: Successful Outgoing External Inter-Cell Handovers CH331:Outgoing External InterCell Handover Commands


×100％

Page13


P-14

Course Name

Performance Index of Outgoing External Inter-Cell Handover(BSC) ZK3176: External Outgoing Cell Handover Success Ratio per BSC ＝

ZCH333:Successful External Outgoing Cell Handovers per BSC ZCH330:External Outgoing Cell Handover Requests per BSC

×100％

External Outgoing Cell Handover Success Ratio per BSC <= External Outgoing Cell Radio Handover Success Ratio per BSC ZK3177: External Outgoing Cell Radio Handover Success Ratio per BSC ＝

ZCH333:Successful External Outgoing Cell Handovers per BSC ZCH331:External Outgoing Cell Handover Commands per BSC

×100％

N

External Outgoing Cell Handover Commands per BSC＝∑ Outgoing External Inter - Cell Handover Commands i =0


Page14


P-15

Course Name

Performance Indices of BSC Overall Handover ZK3180: Handover Success Ratio per BSC ＝ [(ZCH313: Successful Internal Outgoing Cell Handovers per BSC+ ZCH333:Successful External Outgoing Cell Handovers per BSC )/ (ZCH310: Internal Outgoing Cell Handover Requests per BSC + ZCH330:External Outgoing Cell Handover Requests per BSC )]*100%

Handover Success Ratio per BSC <= Radio Handover Success Ratio per BSC ZK3181: Radio Handover Success Ratio per BSC＝ [(ZCH313:Successful Internal Outgoing Cell Handovers per BSC + ZCH333:Successful External Outgoing Cell Handovers per BSC )/ (ZCH311:Internal Outgoing Cell Handover Commands per BSC + ZCH331:External Outgoing Cell Handover Commands per BSC] *100%


Page15


P-16

Course Name

Measurement Points of Inter-MSC Handover MS

BSC-A

MSC-A

HO-Required

MSC-B Prepare_HO

VLR-B

BSC-B

MS

Allocate_HO_NUM Send_HO_Report Send_HO_Report_ACK HO-Request

Prepare_HO_ACK IAI ACM

HO-Command

Process_Access_Signalling

HO-Request-ACK

HO-Access HO-Complete

Clear-Command

Send_End_Signal

Clear-Complete Send_End_Signal_ACK


z

z

z

z

z

Page16

Note: Signaling flow of A interface and Abis interface of inter MSC handover is the same as that of intra MSC handover, only the signaling between two MSC is particular for the inter-MSC handover. As shown in above figure the signaling with "MAP" is of the MAP layer, and signaling of A and Abis interfaces are omitted. After receiving "HO-REQUIRED" of BSC-A (the request message includes CGI of the original cell and target cell), if MSC-A finds that LAC of the target cell doesn’t belong to this MSC, MSC-A will query the "REMOT LAC Table“ (including the LAC and route of the adjacent MSC), and send "Prepare-HO" message to MSC-B according to the route. CGI of the target cell and the indicator of whether to allocate the handover number are included in this Prepare-HO message. According to the received "Prepare-HO" message, if the handover number needs allocating, MSC-B will request the local VLR to allocate the handover number. If VLR has the free handover number, the handover number will be sent to MSC-B through "SendHO-Report". If no handover number is needed, proceed to the next step. After SCCP link between MSC-B and BSC-B is established, MSC-B sends "HO-REQUEST" to BSC-B. After that, BSC-B activates the target cell’s channel, and returns "HO-REQUEST ACK" to MSC-B after receiving the channel activation acknowledgement. According to this message, MSC-B sends "Prepare-HO ACK" to MSC-A. MSC-A establishes the route to MSC-B according to the handover number, and sends "Initialize-Address" (IAI) to MSC-B to help the latter to identify which voice channel is reserved for MS. While MSC-B returns "Address-Complete" (ACM) to MSC-A.


P-17

Course Name



Page17


P-18

Course Name

Analysis of Handover Problem z

Types of handover problems

z

Locating handover problem

z

Causes of handover problem


Page18


P-19

Course Name

Types of Handover Problems z

Types － Possible Results

No handover － Result in call drop

Handover failure － Affect the conversation quality, and call drop

Frequent handover － Affect the conversation quality and increase load of the system


z

Page19

Frequent handover: Handover in GSM is hard handover, it is meaning, before handover to the new channel, MS must release the original channel firstly. So frequent handover will result in word-loss in handover and break-make of conversation, thus affecting the conversation quality.


P-20

Course Name

Method of Locating Handover Problem z

z

Traffic statistics analysis

Handover Failure/Attempt/Success Measurement per BSC

Incoming/Outgoing Internal/External Inter-Cell Handover Measurement per Cell

GSM Cell to GSM Cell Incoming/Outgoing Handover Measurement

Neighbor Cell Level Measurement per Cell

View alarm

Board fault, transmission, clock, etc.

z

Drive test

z

Signaling analysis

Abis interface, A interface, E interface


Page20


P-21

Course Name

Causes of Handover Problem z

Coverage

z

Interference

z

Antenna and feeder system

z

BTS hardware

z

Transmission

z

BSC hardware

z

Data Configuration

z

A interface

z

Target cell congestion

z

Cooperation with equipment of other manufacturers


Page21


P-22

Course Name

Causes of Handover Problem Types

Causes Coverage

No H.O

H.O Failure

Frequent H.O

√

√

√

Interference

√

Antenna and feeder system

√

BTS hardware

√

Transmission

√

BSC hardware

√

Data configuration

√

√

√

A interface

√

√

√


√

√

H.O between equipment of different manufactures


√

Page22


P-23

Course Name

Problem in Coverage and Interference

Coverage

Poor coverage: forest, complicated topography, building direction and indoor coverage

Isolated site: no adjacent cell

Over shooting: island effect result in no adjacent cell

Interference

MS can not access network or receive any signal.


Page23


P-24

Course Name

Island Effect Results in Handover Failure There is no adjacent cell, so handover becomes impossible.

Non-adjacent cell

Adjacent cell N1

Service cell Non-adjacent Adjacent

cell

Cell N2 Non-adjacent

Adjacent

cell

Cell N3

Isolated island resulting from over shooting


z

Page24

In isolated cell coverage area, when MS moves towards the cell edge, the signal becomes weaker and weaker, and since there is no adjacent cell available around, handover cannot be triggered, thus call drop will occur.


P-25

Course Name

Problem in Antenna and Feeder System

High Voltage Standing Wave Ratio (VSWR)

Antenna is not properly installed

Antenna is not parallel

The azimuth and downtilt are not correct

Poor antenna isolation value

RF cables, connectors are loose or incorrect


Page25


P-26

Course Name

Problem in BTS Hardware

DDPU,DCOM, splitter/combiner failure

TRX failure

DTMU failure

Clock failure

Internal communication cable failure


Page26


P-27

Course Name

Problem in Transmission and BSC Hardware z

z

Transmission

Transmission is not stable

Serious BER in transmission

Fault of BSC Hardware

Clock board: the faulty clock board causes clock inconsistency between base stations.


z

z z z

Page27

For BTS clock, BSC clock, and MSC clock, the following standards are used to judge whether they are out of synchronization: MSC: △f/f (frequency deviation) ≤ 1E-8 BSC/BTS: △f/f (frequency deviation) ≤5E-8 Inaccurate clock will cause that MS cannot decode BSIC of the adjacent cell correctly.


P-28

Course Name

Problem in Data Configuration

Unsuitable setting of handover hysteresis and handover priority.

Unsuitable setting of P, N value of statistic time

Unsuitable frequency and adjacent relationship configured in BA1/BA2 table

CGI, BCCH and BSIC in "External Cell Description Table" are different from those in corresponding BSC.

The DPC of BSC in MSC "LAI and GCI Table" is incorrect.


z z

z

Page28

For service cell: Outgoing inter cell handover will fail. For illusory target cell: Incoming inter-cell handover will fail, and successful rate of handover = successful rate of radio handover. Notes: CI can not be configured as "FFFF". "Transmitting BS/MS Power Level": If the measurement report preprocess is enabled, this parameter must be set to "Yes".


P-29

Course Name

Problem in A Interface and Target Cell z

A interface problem

Basically, the insufficient link resource results in the abnormal handover, as well as abnormal communication.

z

Circuit pool numbers are different ,causing the handover failure.


The target cell is congested, which causes the handover failure. Then the target cell should be expanded or reduce its traffic .


Page29


P-30

Course Name

Problem in Cooperation between manufactures z

Handover between equipment of different manufacturers

The signalings at A interface, E interface of the opposite equipment don’t matched with our equipment and can not be recognized or supported, which causes the handover failure, such as voice version, handover number, addressing mode (CGI or LAI).


Page30


P-31

Course Name



Page31


P-32

Course Name

Handover Case 1 z

Fault Description

A 1800 cell of a dual-band network (all the 1800 cells belong to one BSC), the incoming handover success rate of intra-BSC and inter-BSC are low from the beginning of the service, while the outgoing handover success rate of intra-BSC and inter-BSC are normal.


Page32


P-33

Course Name

Handover Case 1 z

Analysis

Register the traffic statistics and analyze whether the low handover success rate is due to the failure of handover from all cells to this cell or from some few cells.

If handover fails from some few cells to this cell, check the handover data and confirm whether there is co-channel and co-BSIC problem.

If handover fails from all other cells to this cell, check the data of this cell.

If data problem is excluded, check the hardware carefully. Check the alarm or perform drive test to locate uplink fault or downlink fault. Check step by step and find out the cause.


Page33


P-34

Course Name

Handover Case 1 z

Solution

Register the incoming inter cell handover measurement function and find that the successful rate of handover from all other cells to this cell is low, although it is not always 0 percent. Via careful data checking, confirm that the data of this cell is correct.

Perform drive test and find that the downlink signal is normal but almost all handovers to this cell are failure. But near the BTS, the handover is successful occasionally, perhaps the problem is with the uplink signal.

Check the uplink channel, antenna, DDPU, there are no problem. Change the TRX, everything is normal.

z

Conclusion

The symptom is that the uplink and downlink at Um interface are unbalanced so uplink voice quality is bad.


Page34


P-35

Course Name

Handover Case 2 z

Fault Description

A subscriber complains that the signal in a place on his way to office is not good. Call drops occurs several times when he drives his car. The phenomenon is that the conversation is suddenly interrupted. The handset signal bar indicates no signal, and then normal again after several seconds.


Page35


P-36

Course Name

Handover Case 2 z

Analysis

Neighbor cell relationship

Isolated island effect

Geographical factors of buildings and mountains result in the weak signal

The signal of a cell suddenly fades drastically due to buildings

BTS equipment or intermittent transmission


z

Page36

The following possibilities can lead to the problem that a call is dropped, that there is no signal during call drop, and that the signal is recovered after several seconds: The most common possibility is that there is no neighbor cell handover relation, so an MS cannot be handed over to the neighbor cell, hence the call drops. After the call drops, the MS re-selects a cell and resides in a good cell, so the signal is recovered. Isolated island caused by cross-coverage also leads to this problem. When a signal exceeds the coverage, an MS uses the signal even when the MS is far away from the cell that provides the signal, but at this time there is no cell that is a neighbor of the cell. When the cell signal becomes weak, handover is impossible, leading to the call drop. After the call drop, the MS re-selects a good cell and displays signal again. At that place, the signal is indeed too weak to maintain conversation due to geographical factors of buildings and mountains. When an MS goes to that place, call drop occurs. After the MS passes the place, the signal is recovered again. In the urban area where there are many BTSs, this problem does not happen. In the suburb area where only one BTS is used for coverage, this problem sometimes occurs. At that place, due to buildings, the signal of a cell suddenly fades drastically, and the MS has not time to hand over, hence the call drop. After the call drop, the MS resides in a good cell again. Signal is interrupted due to the BTS equipment or intermittent transmission.


P-37

Course Name

Handover Case 2 z

Solution

Via DT, it shows that the signal of several cells is

D

strong and is -70 to -80 dBm, so it is unlikely that the problem is caused by bad signal.

3

A

We obtain further information from the subscriber

2

B 3

and know that the driver at that time drive very fast. Therefore it supposed that this problem may occur under high speed movement. Then perform drive test again under high speed for several times and the call drop recurs.


1

C

Page37

The customer service technicians go to the site to perform drive test and discovers that the road from south to north is covered by the signals from four cells, which are cell1 (northward) of BTS C, cell 3 (southwestward) of BTS B, cell 2 (southeastward) of BTS A, and cell 3 (southwestward) of BTS D. In the complained place the signal of several cells is strong and is -70 to -80 dBm. It is unlikely that the problem is caused by bad signal. The drive test shows that the handover relation is normal. The drive route is such that cell 1 of BTS C is first occupied; then the vehicle drives northward and the MS can hand over to cell 3 of BTS B and cell 2 of BTS A. As the vehicle drives on northward, the MS hands over to cell 3 of BTS D. Multiple drive tests along the road show normal result and the problem complained by the subscriber does not occur. To investigate the problem, we obtain further information from the subscriber and know that the driver at that time drove very fast. Therefore we think this problem may occur under high speed movement. We perform drive test again under high speed for several times and the call drop recurs.


P-38

Course Name

Handover Cases 2 z

Solution

Cell 1 of BTS C is far away from cell 3 of BTS D and the two cells are not defined as neighbor cells. Thus an

D

isolated island is generated, handover cannot be performed, and call drop arises. The special thing about this isolated island effect is that it occurs at high

3

A 2

B

speed. Therefore troubleshooting is difficult.

3

After a neighbor relation is added, this problem is solved. On the other hand, because this problem occurs due to insufficient time to hand over, the handover decision time is reduced so that the

1

C

handover can happen timely.


Page38

Cell 1 of BTS C provides coverage along the road and the signal reaches far away. When a vehicle drives very fast, sometimes the MS has not time to hand over to cell 3 of BTS B or cell 2 of BTS A. Cell 1 of BTS C is far away from cell 3 of BTS D and the two cells are not defined as neighbor cells. Thus an isolated island is generated, handover cannot be performed, and call drop arises. The special thing about this isolated island effect is that it occurs at high speed. Therefore troubleshooting is difficult. After a neighbor relation is added, this problem is solved. On the other hand, because this problem occurs due to insufficient time to hand over, the handover decision time is reduced so that the handover can happen timely.


P-39

Course Name

Handover Cases 3 z

Fault Description

This case occurs during the drive test at a new site.

Even when the receive level of serving cell B is higher than the receive level of neighboring cell M, handover is triggered from cell B to cell M.


Page39


P-40

Course Name

Handover Cases 3 z

Analysis

The possible causes of the handover from a cell with a higher level to a cell with a lower level cell are as follows: 1. The PBGT handover threshold for traffic adjustment is lower than 64. 2. The handover is emergency handover due to the heavy load or low quali ty of the original cell with the higher level. 3. The priority of the cell with the higher level is lower than the priority of t he cell with the lower level.


Page40


P-41

Course Name

Handover Cases 3 z

Solution

Analyze the drive test data and find that the downlink receive quality and le vel of the serving cell are normal. Thus, the handover is not associated with the downlink receive quality and level.

Check the data configurations and find that the handover due to heavy loa d is disabled. Thus, the handover is not due to heavy load.

Check the BSC parameter configurations. The result shows that the threshol d for the PBGT handover from cell B to cell M is 68 and that the handover h ysteresis is 4. Thus, the handover is not PBGT handover.


Page41


P-42

Course Name

Handover Cases 3 z

Solution

Check the priority settings of the cells. The result shows that cell B is at level 2 on l ayer 3 and that cell M is at level 1 on layer 3. Check the inter-layer handover threshold and hysteresis. The result shows that the interlayer handover threshold is 20 and that the interlayer handover hysteresis is 5. Thus, when the receive level of cell M is higher than -85 dBm, handover is initiated from cell B to cell M.

Modify the priority level of cell B to level 1. In this way, the cells are at the same pr iority level. The handover problem is solved.


Page42


P-43

Course Name

Handover Cases 4 z

Fault Description

At a suburban area in city A, Huawei BTS3012 M-2 cell, incoming handover success rate lower than 70%, the number of incoming handover failures at busy hours is greater than 35.


Page43


P-44

Course Name

Handover Cases 4 z

Analysis

The hardware such as clock board is checked and no problem is found. Register the traffic measurement of the cell incoming handover, it is found that most of the handover failures occur from the external N-1 cell of the BTS of company S. The field drive test shows that in times of handover failure, there occurs a neighbor frequency in the neighbor cell frequency list. It is suspected that the handover failures are caused by downlink inter-frequency interference. We trace and analyze the handover failure signaling and finds that sometimes the handover failures occur when the MS receives the handover command and sends access pulses to the target cell and sometimes the handover failures occur when the MS reports the SABM frame.


Page44


P-45

Course Name

Handover Cases 4 z

Solution

Therefore the conclusion is that the MS fails to correctly decode the physical message due to the downlink inter-frequency interference, or the MS fails to receive the UA frame from the target cell, resulting in handover failure.

The cell frequency is planned again and is changed from 113 to 121. The problem is solved.


z

Page45

Signaling tracing is very helpful if there are problems in cooperation with other manufacturers’ equipment.


P-46

Course Name

Handover Cases 4 z

Handling Procedure

The BTS hardware alarms are checked.

The cell incoming handover traffic measurement is registered.

Drive test to trace the signaling.

The cell frequency is modified.

z

The traffic measurement is checked and it is confirmed that the number of incoming handover failures decreases and the incoming handover success rate reaches 92%.

Suggestion and Summary

The downlink interference is not represented in the interference band and the range affected by the partial downlink interference is small. We may miss the faulty place during the drive test due to selection of the drive route. It is desirable to perform careful drive test on the abnormal cells.


Page46


P-47

Course Name

Handover Cases 5 z

Fault Description

One BSC6000 has very low HO attempt & during many drive test no HO at all.

KPI indicate that there is very low HO attempt with good success rate.

From Drive test using TEMS, we found out “Missing Neighbor Detection” message appears in neighbor list when MS still try to make HO to serving cell.

The figure below and the corresponding KPI statistics table is presented as below. Corresponding KPI Statistics Table


Page47


P-48

Course Name

Handover Cases 5


Page48


P-49

Course Name

Handover Cases 5 z

Analysis

Neighbors cell configuration is incorrect

Data configuration has some problem, such as “NCC Permitted”


Page49


P-50

Course Name

Handover Cases 5 z

Solution

Compared with other BSC, found out the handover attempts are less than other BSCs even under the same condition.

Checked the neighbor cell configuration, for some cells, there were some neighboring cells missing, but after adding, the problem still existed.

Checked "NCC permitted" , we found that the configuration in the problem BSC is set to be 11110000, that is, enabling NCC 4~7 handover while disabling NCC0~3, but the neighboring cells' NCC is 0~3.

Changed “NCC Permitted” to be 11111111, the problem is solved.


z

z

z

z

z

Page50

To solve the problem of low handover attempts, we firstly compared the handover times with other BSCs under the same scenario and found out the attempts are less than other BSCs even under the same condition, in addition, we also checked the neighbor cell configuration to see whether there are enough neighbors to ensure the handover proper and prompt, for some cells, there were some neighboring cells missing, but after adding, the problem still existed, and we checked the parameters, found out There is a parameter "NCC permitted" relative to the low handover times, the definition of this parameter is as follows: NCC permitted Range: 0 allowed~7 allowed Default: 1111111 Description: Network color code permitted. The value 1 stands for permitted and 0 for forbidden. This parameter is sent in system information 2 and 6. When the cell's NCC is consistent with the value of NCC permitted, then this cell will be measured by MS. And MS will report the measurement report to BTS. This parameter consists of one byte (8bit). Each bit is corresponding to an NCC (0~7) and the last bit is corresponding to NCC 0. If bit N is 0, then MS will not measure the cell level with NCC being N. Note:As MS cannot report the adjacent cell information where NCC is set to 0, the incorrect setting of this parameter will cause MS to be unable to hand over during conversation. See Protocol 0508. This parameter can be used to make MS 's measurements on some adjacent cells optionally. We found that the configuration in the problem BSC is set to be 11110000, that is, enabling NCC 4~7 handover, but from the feedback, the neighboring cells' NCC is 0~3.


P-51

Course Name

Handover Cases 5 z

Suggestions and Summary:

“NCC Permitted” parameter is an important parameter to impact on handover but always been neglected, and if the similar phenomenon happened, this is a clue to solve this problem.


Page51


P-52

Course Name

Handover Cases 6 z

Fault Description

A GSM network, where one MSC is attached with two BSC, is all configured with Huawei equipment. When the two BSC are cutovered, the successful handover rate from BSC1 to BSC2 is 0, but the opposite is normal. Intra-BSC1 and intra-BSC2 handover are normal.

BSC1

0 successful handover rate

BSC2

Normal successful handover rate


Page52


P-53

Course Name

Handover Cases 6 z

Analysis

Register “outgoing inter-cell handover measurement function” of BSC1 and “incoming inter-cell handover measurement function” of BSC2.

Check the data related to handover of BSC1 and BSC2.

Analyze signaling of handover failure.


Page53


P-54

Course Name

Handover Cases 6 z

Solution

Check all data of inter-BSC handover: external cell description data table, BA2 table, CGI of MSC. No problem is found.

Trace A interface message of BSC1 and BSC2. After BSC1 sends "HOREQUIRED", BSC2 has not received "HO-REQUEST" message, but opposite is normal.

The path of data search: MSC goes to "LAI and GCI Table" according to the CGI of the target cell in "HO-REQUIRED" message, and sends HO request to the correct BSC base on the description of the DPC of the cell given in the table.


Page54


P-55

Course Name

Handover Cases 6 z

Solution

The HO-REQUEST message, which should have been sent to BSC2, is sent to BSC1. It is DPC error. After correction, the problem is solved.

z

Conclusion

For the data of inter-BSC handover, besides checking whether CGI at MSC side is consistent with that of opposite and BSC, check whether the DPC is correct. MSC look up target DPC based on the CGI of the target cell. When the DPC is incorrect, "HO-REQUEST" will be sent to wrong BSC.


Page55


P-56

Course Name

Handover Cases 7 z

Fault Description

After all BTSs of a certain BSC32 are swapped to the BSC6000, the PBGT handover proportion decreases from the average value 60% to 40%. The number of bad quality handovers increase. It thinks that the problem is caused by the interference and coverage.

Note:

The PBGT handover proportion is one of the key indexes that the operator uses to evaluate the network coverage quality.

PBGT Handover Proportion = PBGT Handover Requests/Total Handover Requests


Page56


P-57

Course Name

Handover Cases 7 z

Analysis

The possible causes of the decrease in the PBGT handover proportion are as follows:

The number of other handovers with a higher priority than that of the PBGT handover increases. The possible causes are the interference or coverage problems.

The data configurations of PBGT handover change, thus causing an increase in the condition of the PBGT handover and a decrease in the number of PBGT handover attempts.

16-bit value affects the priorities of many neighbor cells with good levels.


z

z

Page57

(1). The PBGT handover algorithm is used to find a cell that meets the following conditions on a real-time basis and determine whether to perform a handover: The cell has less path loss. The level of the neighbor cell is higher than the threshold for the local cell. The cell meets the system requirements. PBGT Handover Proportion = PBGT Handover Requests/Total Handover Requests (2). The possible causes of the decrease in the PBGT handover proportion are as follows: The number of other handovers with a higher priority than that of the PBGT handover increases. The possible causes are the interference or coverage problems. The system parameters that the PBGT handover requires change, thus causing an increase in the number of requirements for the PBGT handover and a decrease in the number of PBGT handover attempts.


P-58

Course Name

z

(3). The requirements of the PBGT handover are as follows: The PBGT handover is implemented only between the cells with the same layer and same level. The priority of the target cell is higher than that of the serving cell. The PBGT handover is triggered only over the TCH. In the 16-bit handover algorithm, other bits with higher priorities than the priority of the level are:

Inter-layer handover threshold adjust bit

Co-BSC/MSC adjust bit

Load adjust bit

Handover layer bit The priorities of these bits are higher than the priority of the level. If they are set improperly, the priorities of many neighbor cells with good levels are lower than the priority of the local cell. Thus, the PBGT handover fails to be triggered.


P-59

Course Name

Handover Cases 7 z

Solution

The problem may be caused by interference and coverage. Check the related traffic measurement. The traffic indexes of the interference band before and after the swap are low. The BTS hardware is operational.

Check the parameters related to the PBGT handover, PBGT Handover Threshold and Decision Time. The parameters of other handovers with higher priories than PBGT handover are proper.

Check the settings of the 16-bit priority. The inter-layer handover threshold and hysteresis are greatly different from those configured on the original BSC. For the existing network, the initial threshold is set to 25 and the hysteresis is set to 3. For the original BSC, the initial threshold is set to 40 and the hysteresis is set to 1.


z

z

z

z

z

z

z

Page59

(1). The problem may be caused by interference and coverage. Check the related traffic measurement. The interference handover proportion before the swap is low. The traffic indexes of the interference band before and after the swap are low. The BTS hardware is operational. The OMC data shows that not all parameters are mapped according to the original network during the swap from the BSC32 to the BSC6000. All BTSs are configured according to the initial template of the BSC6000. (2). Check the parameters related to the PBGT handover, PBGT Handover Threshold and Decision Time. The parameters of the handovers with higher priories than PBGT handover are proper. (3). Check the settings of the 16-bit priority. The inter-layer handover threshold and hysteresis are greatly different from those configured on the original BSC. For the existing network, the initial threshold is set to 25 and the hysteresis is set to 3. For the original BSC, the initial threshold is set to 40 and the hysteresis is set to 1. If the following conditions are met, the priority of the neighbor cell is lower than that of the local cell. Even the level of the neighbor cell is higher than that of the local cell, the PBGT handover fails to be triggered. The threshold for the receiving level is set to 25. The level of the neighbor cell is lower than the value of threshold + hysteresis. The level of the local cell is not lower than the value of threshold - hysteresis. In addition, the sites covered by the BSC are located in the suburb areas. There is a great probability that the receiving level of these sites is 25. Therefore, the settings of 16-bit priority have great impacts on the PBGT handover. After the threshold is set to 40 and the hysteresis is set to 1 in batches on the BSC, the PBGT handover proportion increases to 55%-60%, which is close to the original BSC index value. -82 --88 -71


P-60

Course Name

Handover Cases 7 z


When analyzing the handover traffic measurement, the impacts of the factors in the handover algorithm on the handover priority and handover triggering conditions should be taken into account.

The mapping of all parameters at the cell level is essential to the BSC swap. If parameters are not configured according to the original network, the network indexes may be affected. There are multiple parameters at the cell level.


z

z

Page60

(1). When analyzing the handover traffic measurement, you need to take the impacts of the factors in the handover algorithm on the handover priority and handover triggering conditions into consideration. (2). The mapping of all parameters at the cell level is essential to the BSC swap. If parameters are not configured according to the original network, the network indexes may be affected. There are multiple parameters at the cell level. Therefore, the workload of analyzing and checking the parameters is large if the indexes related to the traffic measurement have problems.


P-61

Course Name

Handover Cases 8 z

Fault Description

Huawei BTSs belong to Huawei BSC independently are inserted into a GSM900 network. After the cutover, the user complains that Huawei BTS handover is not good and that the BSC outgoing handover success rate is below 85%. The drive test shows that the serving cell is Huawei cell A and that then drive toward cell B. The receiving level of cell A is -80 dBm (no downlink power control), while the receiving level of the neighbor cell B of another BSC is -60 dBm. Handover does not happen until MS passes across the cell B. Non-Huawei BSC Huawei BSC

No handover for a long time？ -80dBm

-60dBm

Huawei Cell A Copyright © 2010 Huawei Technologies Co., Ltd. All rights reserved.

Cell B

Page61


P-62

Course Name

Handover Cases 8 z

Analysis

The reason that the receiving level of the neighbor cell is strong enough to prevent handover is as follows:

The BA2 table has the BCCH frequency of cell B but the handover data, including parameters such as external cell CGI, BCCH, BSIC, and "whether co-MSC“ are wrong.

Congestion of cell B results in no handover.

Other parameters are set incorrectly, resulting in no handover.


Page62


P-63

Course Name

Handover Cases 8 z

Solution

Drive test shows that handover does not happen. To verify reasonableness of the data, test MS is used to perform the forced handover. Handover succeeds.

The traffic statistic of the non-Huawei BSCs is checked and confirm that there is no congestion in the target cell B.

Via drive test, it is found that all the new cells of Huawei BSC have this problem. It is suspected that the problem is caused by a common parameter.

The traffic measurement is analyzed and it is found that no PBGT handover happens in the new BSC, while only edge handover happens when the edge handover condition is satisfied, resulting in serious handover delay.


Page63


P-64

Course Name

Handover Cases 8 z

Solution

One of PBGT handover condition is target cell must be the first position in the 16bit criterion. If “Co-BSC/MSC Adj” is set to YES, the cells that do not in the same BSC/MSC are sorted at the last positions. Therefore the PBGT handover of the BSC outgoing handover does not happen. After “Co-BSC/MSC Adj” is set to NO, the problem persists.

The data is further checked and it is found that the default parameter configuration shows that internal cells of Huawei BSC are "layer 3, priority 1" and the external cells are "layer 3, priority 2". Because PBGT handover only happens at the same layer and the same priority, the default configuration does not trigger PBGT handover.

After changing all internal cells priority of Huawei BSC from 1 to 2, it shows that PBGT handover happens by analyzing the traffic measurement. Drive test shows that BSC outgoing handover is not delayed.


Page64


P-65

Course Name

Handover Cases 8 z


Note that the following two default parameters need to be modified in the BTS and BSC insertion plan 1. Set “Co-BSC/MSC Adj” to NO 2. If there is no requirement for traffic balance, set the layer and priority of the internal cells and external cells to the same layer and the same priority.


Page65


P-66

Course Name

Handover Cases 9 z

Fault Description

The drive test of a network in an office shows that in the drive test in a county, there is a strange phenomenon: when the receiving level of the current serving cell (cell A) is as high as –45 dBm, the call still hands over to a cell (cell B) with low level of –78 dBm. Viewed from geographical location, this place should not be covered by cell B. The frequency planning of the network is tight frequency reuse. Therefore the conversation quality becomes worse after handover. Handover occur from high level to low level！ -45dBm

-78dBm

Cell A


Cell B

Page66


P-67

Course Name

Handover Cases 9 z

Analysis

There are the following reasons that can lead to handover from a high-level cell to a low-level cell.

To adjust the traffic, the PBGT handover threshold is set to a value less than 64.

When emergent handover occurs, such as BQ handover (Because the uplink quality is bad for interference, BQ handover occurs).

Though the downlink signal level of the current cell is strong, the uplink signal level is weak and already reaches the handover threshold.

The layer and priority of the low-level cell is higher than those of the high-level cell.


Page67


P-68

Course Name

Handover Cases 9 z

Solution 1. Check BSC data configuration

It is found that the [PBGT HO Threshold]=70 and [Inter-cell HO Hysteresis]=3. Therefore the handover is not a PBGT handover.

2. The drive test data shows that when a handover is triggered, both the downlink receiving quality and the downlink receiving level of MS in the serving cell are good.

This excludes handover caused by downlink receiving quality and level.


Page68


P-69

Course Name

Handover Cases 9 z

Solution 3.

Check the cell priority

It is found that cell B is at layer 3 with priority 1 while cell A is at layer 3 with priority 2. [Inter-layer HO Threshold]=25 and [Inter-layer HO Thysteresis]=5. Therefore when the receiving level of cell B is greater than –80 dBm, handover is triggered. The handover is a better cell handover from a cell in a high receiving level to a cell in a low receiving level. In the initial data configuration, the BSC internal cells are designed at layer 3 with priority 1 and the external cells are at layer 3 with priority 2. The engineer ever changed data during routine maintenance. The priority of cell A is changed to 1, and the priority of cell B is changed to 2, the handover becomes normal.


Page69


P-70

Course Name

Handover Cases 10 z

Fault Description

Huawei mini BTSs and NT BTSs are in mixtured networking, that is to say, few Huawei BTSs are located in a great many NT BTSs.

One day the customer complains that during a drive test, handing over from a Huawei mini BTS to a neighbor cell (NT BTS) is very difficulty, and that the handover request is not sent even when the receiving level of the neighbor cell (NT BTS) is much stronger than that of the serving cell (Huawei mini BTS). Sometimes handover does not happen until the call drop.

All the cells are in the same layer and the same priority. [Co-BSC/MSC Adj] is YES. [Inter-layer HO Threshold]=25, and [Inter-layer HO Thysteresis]=3.


Page70


P-71

Course Name

Handover Cases 10 z

Analysis

Firstly, suspected that the problem of failure to send handover request is caused by improper neighbor cell relationship of the serving cell or by incorrect frequencies in the BA1 table and BA2 table.

The check result shows that the neighbor cell relationship is right and that the frequencies are correct.


Page71


P-72

Course Name

Handover Cases 10 z

Analysis

Secondly, Since [Co-BSC/MSC Adj] is set to YES, compare the 16bit value between serving cell and neighbor cell. Rx_Lev

In the same

14th bit

13th ~ 12th bit

10th ~5th bit

BSC

Serving Cell

>= –88 dBm

/

0

00

0

Serving Cell

< –88 dBm

/

1

00

0

Neighbor Cell

>= –82 dBm

Yes

0

00

0

Neighbor Cell

>= –82 dBm

No

0

01

0

Neighbor Cell

< –82 dBm

Yes or No

1

00

0

①. All the cells are in the same layer and the same priority, and in the same MSC. ②. [Co-BSC/MSC Adj] is YES ③. [Inter-layer HO Threshold]=25(-85dBm), and [Inter-layer HO Thysteresis]=3dB Copyright © 2010 Huawei Technologies Co., Ltd. All rights reserved.

z z

z

z

z

z

z

z

Page72

Since Co-BSC/MSC Adj is set to YES, of the 16bit sequence number, ①.If the serving cell level >= inter-layer handover threshold–hysteresis, which is >=–88 dBm, bit14 is 0, bits 5~10 of the serving cell are the same as those of the neighbor cell, so bits 12, 13 and 14 are all 0. ②.If the serving cell level < inter-layer handover threshold–hysteresis, which is < –88 dBm, bit14 is 1, bits 5~10 of the serving cell are the same as those of the neighbor cell, so bits 12 and 13 are 0, and bit 14 is 1. ③.If neighbor cell level >= inter-layer handover threshold + hysteresis, which is >= –82 dBm, bit 14 is 0, bits 5~10 of the internal cells are the same as those of the serving cell, so bits 12, 13, and 14 of the internal cells are all 0, while bits 5~10 of the external cells are the same as those of the serving cell, so bit 12 is 1, and bits 13 and 14 are 0. ④.If neighbor cell level < inter-layer handover threshold + hysteresis, which is < –82 dBm, bit 14 is 1, bits 5~10, 12, and 13 are all 0, for both internal cells and external cells. ⑤.The actual situation is, when the level of the serving cell is higher than –88 dBm, all the most significant bits of the serving cell are 0. For cells of different BSC, even when the level of the serving cell is higher than –82 dBm, bit 14 is 0. But because bit 12 is 1, the cells of different BSC are still sorted after the serving cell and no normal handover can be initiated (including edge handover, PBGT handover, and inter-layer inter-priority handover). In other words, when CoBSC/MSC Adj is YES and the vicinity of Huawei BTS is BTSs of different BSCs, if the level of the serving cell is not lower than –88 dBm, handover is very difficult to initiate. ⑥.When the level of the serving cell is lower than –88 dBm, bit 14 of the serving cell is 1, bits 5~10, 12, and 13 are 0. If the receiving level of a neighbor cell of different BSC is lower than –82 dBm, bit 14 is 1, bits 5~10, 12, and 13 are 0. In this case all the cells are sorted by the level. If the receiving level of a neighbor cell of different BSC is higher than –82 dBm, bit 14 is 0 and the cell is sorted at the front. That is, if the vicinity of Huawei BTS is BTSs of different BSCs, when the level of the serving cell is lower than –88 dBm, no matter if the level of the neighbor cells is greater than –82 dBm, the cells can be correctly sorted and can initiate normal handover. ⑦.Therefore, if a Huawei BTS is randomly distributed among other BSCs and the vicinity is BTSs of other BSCs, it is recommended to set Co-BSC/MSC Adj to NO.


P-73

Course Name

Handover Cases 10 z

Analysis

Thirdly, the geographical environment around the mini BTS is complex and P/N = 5/4s, respectively.

These values are large. Therefore sometimes when the MS receiving level is lower than –88 dBm, the level decreases so rapidly due to the complex terrain that call drop happens before handover succeeds.

Therefore P/N are changed to 3/2s, respectively. In this case, handover is triggered very quickly at low level.


Page73


P-74

Course Name

Handover Cases 10 z

Solution

The handover performance measurement is registered and it is found that the cell has handover request and that the handover success rate is high. The BSC incoming and outgoing handover performance measurement is registered and handover happens normally between the cell and a NT cell. Drive test shows that when serving cell is Huawei mini BTS and the serving level is below –80 dBm, handover does not happen even when the signal of the neighbor cell is much stronger than that of the serving cell. But when the signal of the serving cell is –90 dBm, sometimes handover happens and succeeds, sometimes due to terrain reason, the level decreases so rapidly that handover does not happen in time and result in call drop at last.


Page74


P-75

Course Name

Handover Cases 10 z

Solution

It is confirmed that all the neighbor cells have configured in the BA1/BA2 table and the corresponding frequency is correct.

The handover-related parameter setting of the cell is configured to be the same as that of other cells. The parameters that affect inter-BSC cell handover are mainly checked and it is found that [Co-BSC/MSC Adj] is set to YES and that P/N time of the edge handover and PBGT handover are 5s/4s, respectively.

The [Co-BSC/MSC Adj] is modified to NO and the P/N time of the edge handover and PBGT handover are changed to 3s/2s, respectively.

DT shows that handover happens when the MS receiving level is high.


Page75


P-76

Course Name

Thank you www.huawei.com


Gsm Handover Problem Analysis

Recommend Documents