TBF Establishment Success Rate(ZTE)

Improvement of TBF Establishment Success Rate R1.1

Improvement of TBF Establishment Success Rate Internal Use Only▲

LEGAL INFORMATION By accepting this certain document of ZTE CORPORATION you agree to the following terms. If you do not agree to the following terms, please notice that you are not allowed to use this document. Copyright © 201 Copyright 2011 1 ZTE CORPORATION. CORPORATION. Any right rights s not expressly expressly grant granted ed herein are reserved. This document contains proprietary information of ZTE CORPORATION. Any reproduction, transfer, distribution, use or disclosure of this document or any portion of this document, in any form by any means, without the prior written consent of ZTE CORPORATION is prohibited. and are regist registered ered trad trademark emarks s of ZTE CORPO CORPORATI RATION. ON. ZTE’s comp company any name, logo and product names referenced herein are either trademarks or registered tradem tra demark arks s of ZTE COR CORPOR PORATI ATION. ON. Oth Other er pro produc ductt and com compan pany y nam names es men mentio tioned ned herein may be trademarks or trade names of their respective owners. Without the prior written consent of ZTE CORPORATION or the third party owner thereof, anyone’s access to this document should not be construed as granting, by implication, estopped or otherwise, any license or right to use any marks appearing in the document. The design of this product complies with requirements of environmental protection and personal security. This product shall be stored, used or discarded in accordance with prod pr oduc uctt ma manu nual al,, re rele leva vant nt co cont ntra ract ct or la laws ws an and d re regu gula lati tion ons s in re rele leva vant nt co coun untr try y (countries). This document is provided “as is” and “as available”. Information contained in this document is subject to continuous update without further notice due to improvement and update of ZTE CORPORATION’s products and technologies.

ZTE CORPORATION Addres s:

Websit e: Email:

NO. 55 Hi-tech Road South ShenZhen P.R.China 518057 http://dms.zte.com.cn http://dms.zte.com.c n (Technical Support) [email protected] [email protected]

ZTE Confidential Proprietary

© 2011 ZTE CORPORATION. All rights reserved.

I


LEGAL INFORMATION By accepting this certain document of ZTE CORPORATION you agree to the following terms. If you do not agree to the following terms, please notice that you are not allowed to use this document. Copyright © 201 Copyright 2011 1 ZTE CORPORATION. CORPORATION. Any right rights s not expressly expressly grant granted ed herein are reserved. This document contains proprietary information of ZTE CORPORATION. Any reproduction, transfer, distribution, use or disclosure of this document or any portion of this document, in any form by any means, without the prior written consent of ZTE CORPORATION is prohibited. and are regist registered ered trad trademark emarks s of ZTE CORPO CORPORATI RATION. ON. ZTE’s comp company any name, logo and product names referenced herein are either trademarks or registered tradem tra demark arks s of ZTE COR CORPOR PORATI ATION. ON. Oth Other er pro produc ductt and com compan pany y nam names es men mentio tioned ned herein may be trademarks or trade names of their respective owners. Without the prior written consent of ZTE CORPORATION or the third party owner thereof, anyone’s access to this document should not be construed as granting, by implication, estopped or otherwise, any license or right to use any marks appearing in the document. The design of this product complies with requirements of environmental protection and personal security. This product shall be stored, used or discarded in accordance with prod pr oduc uctt ma manu nual al,, re rele leva vant nt co cont ntra ract ct or la laws ws an and d re regu gula lati tion ons s in re rele leva vant nt co coun untr try y (countries). This document is provided “as is” and “as available”. Information contained in this document is subject to continuous update without further notice due to improvement and update of ZTE CORPORATION’s products and technologies.

ZTE CORPORATION Addres s:

Websit e: Email:

NO. 55 Hi-tech Road South ShenZhen P.R.China 518057 http://dms.zte.com.cn http://dms.zte.com.c n (Technical Support) [email protected] [email protected]



I


Revision History Product Version

Document Version

Serial Number

Reason for Revision

R1.0

First published

R1.1

The impact of resource configuration on KPIs and relevant statistic measurements and counters are added.

Author Date

Document Version

2010-07-05

R1.0

2010-11-15

R1.1

ZTE Confidential Pr Proprietary

Prepared by

Reviewed by

Approved by

Yuan Shuai

Hou Shuai, and Yang Yong

Zheng Hao and Fei Aiping

Yuan Shuai

Hou Shuai, Yang Yong, and Jiao Qiang

Zheng Hao

© 20 2011 ZT ZTE CO CORPORATION. Al All ri rights reserved.

II


Intended audience: GSM network optimization engineers

Proposal: Before reading this document, you had better have the following knowledge and skills. SEQ 1

Knowledge and skills Null

Reference material Null

2 3

Follow-up document: After reading this document, you may need the following information. SEQ 1

Reference material 3GPP TS 44.060 Protocol

Information Null

2 3



III


About This Document Summary Chapter

Description

1 About TBF

Introduces the concept of TBF.

2 How to Improve the UL/DL TBF Establishment Success Rate

Lists the factors that may influence TBF establishment and gives configuration scheme to improve the UL/DL TBF establishment success rate.

3 UL/DL TBF Establishment Success Rate

Lists formulae about TBF establishment success rate.

4 UL/DL TBF Establishment Process

Lists TBF establishment process.

5 On-site Optimization Cases

Describes a TBF establishment success rate optimization case.

6 Related Measurement and Counters

Lists the counters about TBF establishment success rate.



IV


TABLE OF CONTENTS 1 About TBF..................................................................................................................... 1 2 How to Improve the UL/DL TBF Establishment Success Rate..................................2 3 UL/DL TBF Establishment Success Rate....................................................................8 3.1 UL TBF Establishment Success Rate..........................................................................8 3.2 DL TBF Establishment Success Rate..........................................................................9 3.3 UL GPRS Establishment Success Rate.....................................................................11 3.4 DL GPRS Establishment Success Rate.....................................................................11 3.5 UL EGPRS Establishment Success Rate...................................................................12 3.6 DL EGPRS Establishment Success Rate...................................................................12 4 UL/DL TBF Establishment Process...........................................................................13 4.1 UL TBF Establishment Process.................................................................................13 4.1.1 One-Phase UL TBF Establishment Process on CCCH.......................................13 4.1.2 Two-Phase UL TBF Establishment Process on CCCH.......................................14 4.1.3 UL TBF Establishment Process on PACCH.......................................................15 4.2 DL TBF Establishment Process.................................................................................16 4.2.1 DL TBF Establishment Process on CCCH..........................................................16 4.2.2 DL TBF Establishment Process on PACCH.......................................................17 5 On-site Optimization Cases.......................................................................................18 5.1 A Case of TBF Establishment Success Rate Improvement for a Network in China (1) .........................................................................................................................18 5.1.1 Background Description.....................................................................................18 5.1.2 Cause Analysis and Procedure..........................................................................18 5.1.3 Effect of Parameter Modification.........................................................................19 5.1.4 Verification of the Parameter Modification Effect................................................21 5.1.5 Change of Other Indicators After the Parameter Modification.............................22 5.2 A Case of TBF Establishment Success Rate Improvement for a Network in China (2) .........................................................................................................................23 5.2.1 Background Description.....................................................................................23 5.2.2 Cause Analysis and Procedure..........................................................................23 5.3 A Case of TBF Establishment Success Rate Improvement for an Overseas Network .........................................................................................................................25 5.3.1 Background Description.....................................................................................25 5.3.2 Cause Analysis and Procedure..........................................................................25 6 Related Measurement and Counters.........................................................................28 6.1 PS Basic Measurement..............................................................................................28 6.2 Usage of PS Resources.............................................................................................30 6.3 UL/DL GPRS/EDGE Data TBF ............................................................................. ....31



V


1

About TBF The temporary block flow (TBF) is one of the most important concepts of the packet service. It is a physical connection used by two radio resource (RR) entities to transfer LLC PDUS in one direction on packet data radio channels. The TBF is allocated radio resources on one or more PDCHs and comprises a number of RLC/MAC blocks carrying one or more LLC PDUs. TBF is temporary and is maintained only for the duration of the data transfer (that is, until there are no more RLC/MAC blocks to be transmitted and, in RLC acknowledged mode, all the transmitted RLC/MAC blocks have been successfully acknowledged by the receiver). Each TBF is assigned with a temporary flow identity (TFI) by the network. The TFI acts as the MS identifier on the RLC/MAC layer. And one TFI can be used for transmissions in different directions (uplink or downlink). The TFI is assigned in the resource assignment message and is sent before the LLC frame of a TBF is sent. An RLC/MAC block associated with a certain TBF must comprise a TFI. If an RLC data block is transmitted, the TBF is identified by the TFI and the direction (uplink or downlink) in which the RLC data block is sent. If an RLC/MAC control message is transmitted, apart from the TFI, the transmission direction and the message type are also used to identify the TBF. In the GPRS/EGPRS KPI system, UL/DL TBF establishment success rate is a quite important indicator. Whether the TBF can be successfully established determines whether data services can be provided normally. From the view of users’ perception, TBF establishment success rate determines whether users can logon the Internet and whether they can surf the internet fluently. In a word, TBF establishment success rate is the basis of GPRS/EDGE data transmission.



1


2

How to Improve the UL/DL Establishment Success Rate

TBF

To improve the UL/DL TBF establishment success rate, we need to start with the TBF establishment process. There are many factors that influence the TBF establishment success rate, such as PCU processing capability, Abis transmission band width, the number of PDCHs, and the transmission BER on the Um interface and the Abis interface. These factors will influence the whole process of TBF establishment. This article introduces all the factors that may influence TBF establishment with the analysis of the TBF establishment process. And it also gives suggestions on how to configure relevant parameters to improve the UL/DL TBF establishment success rate. 1.

The TBF is fundamental for the performance of all the data services. To ensure good Internet experience for users, sufficient network resources should be guaranteed. The allocation of Abis resource and DSP resource has a great influence on TBF establishment. It is recommended to plan the network resources in advance. For the method of resources allocation, see the relevant guide. To optimize the running network, you can check the current configuration according to the above guide, or observe the Abis resource and DSP resource congestions in the running network by checking the statistic measurement of the GPRS UL/DL data TBF establishment and that of the EGPRS UL/DL data TBF establishment. And then, you can make some adjustment based on these statistics.

2.

The amount of channel resources also influences TBF establishment. When the average number of users and the maximum number of users per unit time exceed the limit allowed by the network, some users may not be able to access the network due to insufficient resources. This will impact users’ perception. In this case, you need to take statistics of the average number of EDGE/GPRS users, the maximum number of EDGE/GPRS users, and the maximum number of accessed users per timeslot. And then, adjust the configuration of channel numbers. For the setting of SDR PDCHs, see Special Subject Manual for GSM NPO---SDR Channel Configuration.

3.

PDCH configuration also influences the UL/DL TBF establishment success rate. Theoretically, this is not a problem. In practice, the PDCH is better to be configured on the BCCH than on the TCH. Usually, the multiplexing rate of the TCH frequency is higher than that of the BCCH frequency, and on average, the interference on the TCH is greater, which will impact TBF establishment. Based on statistics, the TBF establishment success rate of the PDCH configured on the BCCH is 1%–2% higher than that of the PDCH configured on the TCH.

4.

The adjustment of the access mode also influences TBF establishment. There are two ways to access the network: two-phase access and one-phase access. The greatest difference between them is as follows. When two-phase access is



2


adopted, after the network receives CHANNEL REQUEST from the MS, it will allocate a single block to the MS. And on this single block, the MS will send PACKET RESOURCE REQUEST, which carries some MS request information. According to the requirement of the MS, the network will allocate resources to it. Only when the network receives PACKET RESOURCE REQUEST sent by the MS on the single block, can two-phase access be successful. Because there is just one chance for the network to receive this message, if the UL BLER is too high, or the message is lost during Abis transmission, it is very likely that the BSC will fail to receive this single block. For one-phase access, the BSC has more than one chance to receive the UL block. Even if it fails to receive the UL block for one time, the MS will send radio blocks for many times. As long as the BSC receives the UL block during a certain period (T3166, 5S) or within a certain number of blocks (N3104 (3 × BsCvMax + 9) × number of channels), TBF link can be successfully established. When the radio environment is not desirable, one-phase access is far easier than two-phase access. Therefore, for field optimization, if two-phase access is changed to two-phase access not compulsory, the UL TBF establishment success rate will be greatly increased. 5.

The optimization of T3168 also influences TBF establishment. T3168 is used to set the length of time when the MS waits for the PACKET UPLINK ASSIGNMENT message. After the MS sends PACKET RESOURCE REQUEST, or after the MS sends PACKET CONTROL ACKNOWLEDGEMENT to apply a new TBF, it begins to wait for PACKET UPLINK ASSIGNMENT. And T3168 defines when the MS should stop waiting for PACKET UPLINK ASSIGNMENT. After T3168 times out, the MS will restart packet access process. If the MS sends PACKET RESOURCE REQUEST for four times, it will indicate that TBF establishment fails and will report RLC/MAC error to the upper layer. The value of T3168 will influence the TBF establishment success rate. If the value of T3168 is small, the allowed time for TBF establishment is short. When the radio environment is bad, the TBF establishment success rate will be low. On the contrary, if the value is large, the MS will have more time to judge whether TBF establishment fails. However, because the packet access lasts longer, the system performance will deteriorate. Therefore, when the radio environment is bad, the value of T3168 should be set larger to increase TBF establishment success rate.

6.

If the coding scheme (CS) is too high, while data services are carried out, UL TBF should be established during DL transmission. In this case, if the uplink is interfered or the UL RxLev is poor, improper coding adjustment will cause UL data transmission failure. And TBF establishment will be impacted. Therefore, it is recommended to adopt CS1 or CS2 at most.

7.

Balance between the uplink and downlink. When uplink and downlink RxLevs are unbalanced, on the edge of a cell, either uplink or downlink data cannot be received. This will cause TBF establishment failure. To find out whether the uplink and downlink is balanced, you can check the BTS transmit power first to see whether it is consistent with that of the current network and whether it is too strong. And then you can check the tower amplifier, BTS amplifier, antenna interface, and



3


other relevant parts that may influence UL/DL RxLev. In traffic statistics, unbalance problem is usually indicated by big difference between average UL RxLev and DL RxLev, poor UL or DL RxQual (high proportion of RQ5, RQ6, and RQ7), and low immediate allocation success rate or assignment success rate. If there is unbalance problem in the network, you should optimize the coverage of UL/DL signals. 8.

Polling retry times also influences TBF establishment. Polling is sent to the MS from the network side. If the MS gives no response, the system will send Polling repeatedly. Until the number of retry times reaches the maximum value, the system will release the MS instance. When the radio environment is not desirable, it is recommended to set Polling retry times to 7 or above. Remarks: Large traffic on the GB interface will cause time delay and increase the number of Polling failures to the MS, which will also impact the TBF establishment success rate (2%).

9.

Maximum blocks transmitted in each TS (BsCvMax), which is a parameter used at the MS side. It is a GPRS cell option parameter and is broadcast to the MS in the PSI1, PSI13 and SI13 messages, indicating the value of the maximum blocks (BsCvMax) sent in each timeslot. This parameter decides the value of T3198 that the MS should use when acting as the sending party, the value of T3200 used by the MS in non-DRX mode (= 4 x the time represented by BsCvMax), and the value of N3104max (=3 x (BsCvMax + 3) x Number of timeslots assigned on the uplink). All the uplink data blocks sent by the MS contain the field of countdown value (CV), and the network can use this field to calculate the number of data blocks to be sent on the uplink TBF. In the setting of this parameter, BsCvMax, T3198, and N3104max should be taken into consideration. For BsCvMax, if the extended TBF is not enabled, new data arrives after the countdown starts can only be transmitted on a new TBF link. If the extended TBF is enabled, when new data arrives after the countdown starts, CV will count again, so that the new data can be transmitted over the current TBF. Therefore, if this parameter value is too large, when the MS that does not support extended UL TBF transmits UL data discontinuously, it has to wait until TBF is released after CV is 0 and UL TBF is reestablished, so that new data can be transmitted. If this parameter value is too small, the TBF establishment success rate will be impacted. Therefore, proper setting of this parameter will help increase the average transmission rate. As for T3198, proper setting of the time spent in waiting for PACKET ACK/NACK will facilitate window switch in the PS service. For N3104max, when N3104 reaches N3104max, the packet access process will restart. If the preemptive judgment is not completed after the packet access process restarts for four times, TBF establishment is regarded to be failed. Therefore, proper setting of N3104max can help reduce preemptive judgment time and facilitate the MS to restart the packet access process. For GPRS, this parameter is recommended to be 6; and for EDGE, 12 is recommended.

10. DrxTimerMax optimization also influences TBF establishment. The maximum value of the DRX timer means the longest time when the MS can work in non-DRX mode when the MS changes from packet transmission mode to the packet idle mode. After the MS changes from the packet transmission mode to the packet idle



4


mode, it will stay in non-DRX mode for a while. When TBF is released and the MS is in non-DRX mode, the MS will monitor all the CCCH blocks, and the PCU will keep the MS related context. How long the MS related context will be kept is determined by the minimum value of DrxTimerMax and NonDrxTimer . When the network is not configured with the PCCCH, in non-DRX mode, IMMEDIATE ASSIGHMENT COMMAND can be transmitted on the PCH and AGCH in about 40 ms, which is much shorter than the time needed to transmit this command in DRX mode. Therefore, during the period of non-DRX mode, TBF establishment time is short. In DRX mode, the MS can only monitor paging messages in the home paging group, and receive IMMEDIATE ASSIGNMENT MESSAGE on all the paging blocks and AGCH reserved blocks. And the time for the MS to receive the paging message is long (on average, about half of the paging multiframe cycle). If BS_PA_MFRMS is 2, the receiving period is about 235 ms. However in non-DRX mode, MS consumes more power. Therefore, if the value of DrxTimerMax is large, TBF establishment time can be shortened, but the MS consumes more power. If this value is small, the MS consumes less power; however, in DRX mode, because the addition of paging process increases the signaling load for the system, data transmission delay will increase. In the current network, DrxTimerMax is 2 s by default. It is suggested to modify it to 4 s. 11. NcNoDrxPer is a measurement parameter for network control, and is used at the MS side. This parameter is broadcasted to the MS in the PSI5 message, indicating the minimal time for the MS to stay in the non-DRX mode after sending an NC measurement report. The default value of NC is 2 (0.48 s). It is recommended to modify it to 4 (0.96 s). 12. The setting of N3105. During the process of packet DL transmission, BSS will set the RRBP field on the DL RLC data block at a certain interval to notify the MS to send the RLC/MAC CONTROL message on the relevant UL block. For a TBF, if the RLC/MAC CONTROL message on the designated UL block is lost for over N3105max times, T3195 will start. When T3195 times out, the network can use the TFI resource again. The value of this parameter can be set to 10. When the field radio environment is bad, the value can be larger (30 –50 is recommended). 13. PFCSupport enabled or disabled. This parameter determines whether the BSS supports PFC function. The PFC function is realized by the joint work of the MS and SGSN. The BSS supports the PFC process by the setting of BSC global parameters. Therefore, the PFC function is valid only when the BSC global parameter PFCSupport is enabled. It is recommended to enable this function. The sites that do not need this function can adopt the default value (disabled). 14. Abis resource allocation parameters: PsAbisThs, RS, RR, RA, RM, RB, RG, and RE

−

PsAbisThs refers to the use proportion of the PS channel provided by the fully dynamic Abis resource pool. The value of this parameter indicates the maximum Abis resources that the system can provide. The default value is



5


recommended.

−

RS, the weight of UL throughput on the Gb interface is a dynamic Abis adjustment parameter. The default value is 12 and the recommended value is 20.

−

RR, the weight of DL throughput on the Gb interface is a dynamic Abis adjustment parameter. The default value is 12 and the recommended value is 20.

−

RA, the weight of number of EGPRS MS access requests is a dynamic Abis adjustment parameter. The default value is 6 and the recommended value is 10.

−

RM, the weight of maximum number of bandwidth requests is a dynamic Abis adjustment parameter. The default value is 14 and the recommended value is 25.

−

RB, the weight of maximum bandwidth usage is a dynamic Abis adjustment parameter. The default value is 6 and the recommended value is 10.

−

RG defines the proportion of GPRS frequency carrier. Its default value is 15, and the recommended value is 5.

−

RE defines the proportion of EDGE frequency carrier. Its default value is 35, and recommended value is 10.

It is required that the sum of these seven parameters ( RS, RR, RA, RM, RB, RG, RE) should be 100. 15. Highest coding scheme of signaling TBF and GPRS/EDGE initial coding scheme. Valid bytes of the signaling block are very few; therefore, it is unnecessary to use high CS to transmit blocks. Using lower CS to transmit blocks can improve TBF establishment success rate, transmission success rate, throughput, and other relevant indicators. And the TBF abnormal failure rate can also be reduced. It is recommended that the CS of the signaling TBF is not higher than MCS6. GRPS initial CS is set to CS2. And EDGE initial CS should be set according to the field radio environment. MCS4 or lower is recommended. 16. The EGPRS packet channel request access program support parameter should be set to Yes, if the EDGE service is enabled; and it should be set to No, if the GPRS service is enabled. If two-phase access is not compulsory, the UL access mode is determined by the MS. However, the field experience shows that with this function enabled, the proportion of one-phase access is increased. Because the one-phase access success rate is higher than the two-phase access, for the network that supports EGPRS, with this function enabled, the UL TBF establishment success rate can be improved.



6


17. USFGranularity . It is assigned to the MS in the PACKET UPLINK ASSIGNMENT message, indicating the uplink block allocation granularity of the GPRS MS from the network when the cell employs the media control mode of dynamic allocation. 0 means that the UL block allocation granularity is 1, and 1 means that the UL block allocation granularity is 4. If the value is small, resources can be saved and resource allocation can be more flexible. If this value is set to 4, but there are no four continuous granularities, allocation will fail. Therefore, it is recommended to set this parameter to 0 on site.



7


3

UL/DL Rate

TBF

Establishment

3.1

UL TBF Establishment Success Rate UL TBF establishment success rate = Number establishments/Number of UL TBF establishment attempts KPI KPI Definition

of

Success

successful

UL

TBF

UL TBF establishment success Rate Formula 1: (Number of GPRS UL signaling TBF establishments + Number of EGPRS UL signaling TBF establishments + Number of GPRS UL data TBF establishments + Number of EGPRS UL data TBF establishments) × 100%/(Number of GPRS UL TBF establishments in release state + Number of GPRS UL TBF establishments with existing UL resource + Number of GPRS UL TBF establishment requests (PACCCH) + Number of GPRS UL TBF establishment requests (CCCH/PCCCH) + Number of EGPRS UL TBF establishment requests in release state + Number of EGPRSUL TBF establishment requests with existing UL resource + Number of EGPRS UL TBF establishment requests (PACCCH) + EGPRS UL TBF establishment requests (CCCH/PCCCH)) ZTE system did not count resources allocations and data retransmissions after delay state into link establishment. From now on, they should be counted as in other vendors, such as NSN. Formula 2: (Number of GPRS UL signaling TBF establishments + Number of EGPRS UL signaling TBF establishments + Number of GPRS UL data TBF establishments + Number of EGPRS UL data TBF establishments + Number of GPRS UL TBF resource reallocations due to LLC transmission + Number of EGPRS UL TBF resource reallocations due to LLC transmission + Number of GPRS UL TBF resource reallocations due to DL TBF establishment + Number of EGPRS UL TBF resource reallocations due to DL TBF establishment + Number of GPRS UL TBF resumptions in extend uplink state + Number of EGPRS UL TBF resumptions in extend uplink state) × 100%/(Number of GPRS UL TBF establishments in release state + Number of GPRS UL TBF establishments with existing UL resource + Number of GPRS UL TBF establishment requests on PACCH + Number of GPRS UL TBF establishment requests on CCCH/PCCCH + Number of EGPRS UL TBF establishments in release state + Number of EGPRS UL TBF establishments with existing UL resource + Number of EGPRS UL TBF establishment requests on PACCH + Number of EGPRS UL TBF establishment requests on CCCH/PCCCH + Number of GPRS UL TBF resource reallocation requests due to LLC transmission + Number of EGPRS UL TBF resource reallocation requests due to LLC transmission + Number of GPRS UL TBF resource reallocation requests due to DL TBF establishment + Number of EGPRS UL TBF



8


resource reallocation requests due to DL TBF establishment + Number of GPRS UL TBF resumptions in extend uplink state + Number of EGPRS UL TBF resumptions in extended uplink state) Formula 1: (C900040025 + C900040033 + C900040026 + C900040034)/(C900040159 + C900040160 + C900040161 + C900040168 + C900040163 + C900040164 + C900040165 + C900040166)

KPI Formula

V6.1

V6.2

Notes

3.2

Because the resource reallocation and data resending after delay are not counted as link establishments in ZTE system, the statistics of the two items can be made (Other vendors, such as NSN makes the statistics in this way). Formula 2: (C900040025 + C900040033 + C900040026 + C900040034 + (C900040124 C900040125) + (C900040132 - C900040133) + (C900040122 C90004000123) + (C900040130 - C900040131) + C901010001 + C901010022)/(C900040159 + C900040160 + C900040161 + C900040168 + C900040163 + C900040164 + C900040165 + C900040166 + C900040124 + C90004000132 + C90004000122 + C90004000130 + C901010001 + C901010022 ) The same as the above

Formula 2 is recommended, because it has been used in the Hongkong CSL project, and its KPI definition is consistent with that of other manufacturer’s formula. Enable the one-phase access function, which supports EGPRS_Channel_Req , to improve this KPI. If the function is not enabled, the value of this KPI should be 2% lower. This KPI formula is quoted from GSM RNO Subject-PS KPI_R1.2 . The field engineer should pay attention to the update time of this document and refer to the latest one.

DL TBF Establishment Success Rate DL TBF establishment success rate = Number establishments/Number of DL TBF establishment attempts KPI KPI Definition

of

successful

DL

TBF

DL TBF establishment success Rate Formula 1: (Number of GPRS DL signaling TBF establishments + Number of EGPRS DL signaling TBF establishments + Number of GPRS DL data TBF establishments + Number of EGPRS DL data TBF establishments) × 100%/(Number of GPRS DL TBF establishment requests before T1392 expiration + Number of GPRS DL TBF establishments using DL TBF resource + Number of GPRS DL TBF establishment requests on PACCH + Number of GPRS DL TBF establishment requests on CCCH/PCCCH + Number of EGPRS DL TBF establishment requests before T1392 expiration + Number of EGPRS DL TBF establishments using existing DL TBF resource + Number of EGPRS DL TBF



9


establishment requests on PACCH + Number of EGPRS DL TBF establishment requests on CCCH/PCCCH) ZTE system did not count resources allocations and data retransmissions after delay state into link establishments. From now on, they should be counted as in other vendors, such as NSN. Formula 2: (Number of GPRS DL signaling TBF establishments + Number of EGPRS DL signaling TBF establishments + Number of GPRS DL data TBF establishments + Number of EGPRS DL data TBF establishments + Number of GPRS DL TBF resource reallocation requests due to UL TBF establishment + Number of EGPRS DL TBF resource reallocation requests due to UL TBF establishment + Number of GPRS DL TBF resource reallocation requests due to LLC transmission + Number of EGPRS DL TBF resource reallocation failures due to LLC transmission + Number of GPRS DL TBF resumptions in delay state + of EGPRS DL TBF resumptions in delay state) × 100%/(Number of GPRS DL TBF establishment requests before T1392 expiration + Number of GPRS DL TBF establishments using DL TBF resource + Number of GPRS DL TBF establishment requests on PACCH + Number of GPRS DL TBF establishment requests on CCCH/PCCCH + Number of EGPRS DL TBF establishment requests before T1392 expiration + Number of EGPRS DL TBF establishments using existing DL TBF resource + Number of EGPRS DL TBF establishment requests on PACCH + Number of EGPRS DL TBF establishment requests on CCCH/PCCCH + Number of GPRS DL TBF resource reallocation requests due to UL TBF establishment + Number of EGPRS DL TBF resource reallocation requests due to UL TBF establishment + Number of GPRS DL TBF resource reallocation requests due to LLC transmission + Number of EGPRS DL TBF resource reallocation failures due to LLC transmission + Number of GPRS DL TBF resumptions in delay state + Number of EGPRS DL TBF resumptions in delay state) In some areas, high DL success rate is required. For example, for some networks in Indonesia, DL resource allocation success rate is adopted as an indicator of DL TBF success rate. This formula, with the indicator of 99% or above is not correct. Formula 3: 1 – (Number of GPRS DL TBF establishment failures due to database allocation failure + Number of EGPRS DL TBF establishment failures due to database allocation failure)/(Number of GPRS DL TBF establishment requests before T1392 expiration + Number of GPRS DL TBF establishments using DL TBF resource + Number of GPRS DL TBF establishment requests on PACCH + Number of GPRS DL TBF establishment requests on CCCH/PCCCH + Number of EGPRS DL TBF establishment requests before T1392 expiration + Number of EGPRS DL TBF establishments using existing DL TBF resource + Number of EGPRS DL TBF establishment requests on PACCH + Number of EGPRS DL TBF establishment requests on CCCH/PCCCH) KPI Formula

V6.1


Formula 1: (C900040007 + C900040015 + C900040008 + C900040016)/(C900040141 + C900040142 + C900040143 + C900040144 + C900040145 + C900040146 + C900040147 + C900040148)


10


Formula 2: (C900040007 + C900040015 + C900040008 + C900040016 + (C900040110 C900040111) + (C900040116 C900040117) + (C900040112 C900040113) + (C900040118 - C900040119) + C901000001 + C901000023)/(C900040141 + C900040142 + C900040143 + C900040144 + C900040145 + C900040146 + C900040147 + C900040148 + C900040110 + C900040116 + C900040112 + C900040118 + C901000001 + C901000023) Formula 3: 1 - (C900040109 + C900040115)/(C900040141 + C900040142 + C900040143 + C900040144 + C900040145 + C900040146 + C900040147 + C900040148) V6.2

Notes

3.3

Formula 2 is recommended, because it has been used in the Hongkong CSL project, and its KPI definition is consistent with that of other manufacturer’s formula. If the network requires this KPI to be larger than 99%, Formula 2 is recommended (see KPI Definition) This KPI formula is quoted from GSM RNO Subject-PS KPI_R1.2 . The field engineer should pay attention to the update time of this document and refer to the latest one.

UL GPRS Establishment Success Rate KPI KPI Definition KPI Formula

Notes

3.4

The same as the above

UL GPRS establishment success Rate This KPI cannot be obtained. See Notes. V6.1 V6.2 In the UL establishment process, only after the Packet Resource Request message is received, GPRS or EGPRS can be determined. Therefore, it is impossible to obtain GPRS/EGPRS establishment success rate independently in the uplink.

DL GPRS Establishment Success Rate KPI

KPI Definition

KPI Formula

DL GPRS establishment success Rate (Number of GPRS DL signaling TBF establishments + Number of GPRS DL data TBF establishments) × 100%/(Number of GPRS DL TBF establishment requests before T1392 expiration + Number of GPRS DL TBF establishments using DL TBF resource + Number of GPRS DL TBF establishments using DL TBF resource + Number of GPRS DL TBF establishment requests on PACCH) V6.1


(C900040007 + C900040008)/(C900040141 + C900040142 + C900040143 + C900040144)


11


V6.2 Notes

3.5

Because the UL GPRS establishment rate cannot be counted, the independent DL GPRS TBF establishment success rate is not recommended.

UL EGPRS Establishment Success Rate KPI

UL EGPRS establishment success Rate

KPI Definition

This KPI cannot be obtained. See Notes.

KPI Formula

Notes

3.6


V6.1 V6.2 In the UL establishment process, only after the Packet Resource Request message is received, GPRS or EGPRS can be determined. Therefore, it is impossible to obtain GPRS/EGPRS establishment success rate independently in the uplink.

DL EGPRS Establishment Success Rate KPI

KPI Definition

KPI Formula

Notes

DL EGPRS establishment success Rate (Number of EGPRS DL signaling TBF establishments + Number of EGPRS DL data TBF establishments) × 100%/(Number of EGPRS DL TBF establishment requests before T1392 expiration + Number of EGPRS DL TBF establishments using existing DL TBF resource + Number of EGPRS DL TBF establishment requests on PACCH + Number of EGPRS DL TBF establishment requests on CCCH/PCCCH) V6.1

(C900040015 + C900040016)/(C900040145 + C900040146 + C900040147 + C900040148)

V6.2


Because the UL EGPRS establishment rate cannot be obtained, independent DL EGPRS TBF establishment success rate is not recommended.



12


4

UL/DL TBF Establishment Process

4.1

UL TBF Establishment Process

4.1.1

One-Phase UL TBF Establishment Process on CCCH BSS

MS

RACH/PRACH

(Packet) CHANNEL REQUEST

IMM Assign/Packet Uplink Ass AGCH/PAGCH PDTCH PACCH PDTCH PACCH PDTCH

RLC DATA BLOCK( TLLI) Packet Uplink Ack /Nack(TLLI) RLC DATA BLOCK(TLLI) Packet Uplink Ack /Nack(TLLI) RLC DATA BLOCK(Without TLLI)

......

1.

After the MS sends CHANNEL REQUEST or PACKET CHANNEL REQUEST on the RACH/PRACH at the predetermined time and for a certain number of times, it will leave packet idle mode. If the PBCCH is configured, the MS will send messages on the PRACH; otherwise, on the RACH.

2.

When the BSS at the network side receives CHANNEL REQUEST or PACKET CHANNEL REQUEST on the CCH/PCCCH channel, it will send IMMEDIATE ASSIGNMENT or PACKET UPLINK ASS in unacknowledged mode to allocate UL resources for the MS.

3.

For the one-phase access, after the MS receives IMMEDIATE ASSIGNMENT or PACKET UPLINK ASS, the preemptive judgment mechanism is as follows: 1.

After the MS receives the UL assignment message, the first three RLC data blocks sent by the MS must carry TLLI. When the network receives the first RLC data block, it will response it with the PACKET UPLINK ACK message,



13


and the received TLLI must be included in the message. Then the MS will receive the PACKET UPLINK ACK message. If the included TLLI is identical with that of the MS, the radio resource is allocated to this MS. Otherwise, the MS should exit. 2.

4.1.2

When the BBS at the network side receives the first RLC data block (TLLI) correctly, the preemptive judgment of the one-phase access is completed at the network side.

Two-Phase UL TBF Establishment Process on CCCH BSS

MS

RACH/PRACH

(Packet) CHANNEL REQUEST

IMM ASSIGN/Packet Uplink Ass AGCH/PAGCH PACCH

Packet Resource Request (with TLLI) Packet Uplink Assign(with TLLI)

PACCH PDTCH PACCH PDTCH

RLC DATA BLOCK Packet Uplink Ack /Nack RLC DATA BLOCK

......

1.

During the two-phase access, the IMM ASSIGN or PACKET UPLINK ASS message will allocate a PACCH single block to the MS. With this single block, the MS sends the PACKET RESOURCE REQUEST message (with TLLI) to the network.

2.

According to the information of PACKET RESOURCE REQUEST, the network will send PACKET UPLINK ASSIGN (with TLLI) to the MS to allocate specific channel resources.

3.

The preemptive judgement is as follows: When the MS receives the second PACKET UPLINK ASSIGN, it will compare whether the TLLI included in this message is consistent with that included in PACKET RESOURCE REQ sent by the MS. If it is, the preemption succeeds.



14


4.1.3

UL TBF Establishment Process on PACCH MS

PDTCH PACCH PACCH PDTCH PACCH PDTCH

BSS

RLC DATA BLOCK Packet Downlink Ack/Nack Channel request Packet Uplink Ass ignment RLC DATA BLOCK Packet Uplink Ack /Nack RLC DATA BLOCK

......

1.

During DL TBF period, the MS sends PACKET DOWNLINK ACK/NACK including descriptive information about the channel request on the PACCH to launch the packet access process.

2.

The BSS sends PACKET UPLINK ASSIGNMENT or PACKET TIMESLOT RECONFIGURE on the PACCH to allocate resources to the MS.

3.

The MS transfers to the assigned UL PDCH and starts sending RLC data blocks (without TLLI).



15


4.2

DL TBF Establishment Process

4.2.1

DL TBF Establishment Process on CCCH MS

AGCH/PAGCH PACCH PACCH

BSS

IMM ASSIGN/Packet Downlink Ass Packet Polling request (With TFI) Packet Control Acknowledgement (AB) Packet Power Ctrl/Timing Advance

PACCH TIMING ADVANCE

PACCH PDTCH

RLC DATA BLOCK

......

1.

If there is a request to send LLC PDU to the MS in packet idle mode at the BBS side, the DL TBF establishment process is triggered.

2.

The network sends the DL assignment message PACKET DOWNLINK ASS or IMM ASSIGN on the PAGCH or AGCH block of the MS. If the PBCCH is configured, the network will send messages on the PAGCH; otherwise, on the AGCH.

3.

The DL assignment message should include timing advance. If the BSS does not have valid initial timing advance (TA), it should send PACKET POLLING REQUEST to notify the MS to send PACKET CONTROL ACKNOWLEDGEMENT, so that it can obtain timing advance.

4.

After the MS receives the above message, it will send four burst pulses PACKET CONTROL ACK. And the BSS uses these four access burst pulses to calculate the initial timing advance of the MS.

5.

The network sends PACKET POWER CONTROL/TIMING ADVANCE to provide the MS with TA. If the assignment message contains no TA value, before the MS receives a valid TA, normal burst is not allowed to send on the UL link (such as the PACKET DOWNLINKACK/NACK message).



16


4.2.2

DL TBF Establishment Process on PACCH MS

BSS

...... PDTCH PACCH

PDTCH

RLC DATA BLOCK Packet Downlink Ass ignment Or Packet Timeslot Reconfigure RLC DATA BLOCK

......

1.

During UL transmission, The BSS at the network side sends PACKET DOWNLINK ASSIGNMENT or PACKET TIMESLOT RECONFIGURE through the PACCH to the MS, so as to launch DL TBF establishment. If PACKET TIMESLOT RECONFIGURE is sent, this message should include DOWNLINK_TFI_ASSIGNMENT field. And the MS multi-timeslot limit must be observed.

2.

After the MS receives the assignment message and after the TBF start time (if provided), the MS should transfer to the assigned PDCH to transmit the DL data.



17


5

On-site Optimization Cases

5.1

A Case of TBF Establishment Success Rate Improvement for a Network in China (1)

5.1.1

Background Description ZTE’S GPRS optimization engineer checked the GPRS performance data of the operator’s current network and found that the UL/DL TBF establishment success rates of some BSCs were not high. The performance data of the whole network during busy hours at night is shown in the following table. UL TBF Esta blish ment Succ ess Rate (%)

Start Time

Comple tion time

Number of UL TBF Establish ment Requests (UL Resource Available )

200811-16

2008-1117

19976048

6269943

11559545

10739971

25846888

96.78 %

200811-17

2008-1118

19671183

6031754

11152743

11510653

24220614

96.95 %

Number Number of UL of UL TBF TBF Establis Establish hment ment Reques Requests ts (on (on PACCH) CCCH)

Number of UL Signaling TBF Establish ment Times

Number of UL Data TBF Establish ment Times

From the above table, we can see that the UL TBF establishment rate of the whole network is about 96% –97%, which is far from our optimization target.

5.1.2

Cause Analysis and Procedure After the analysis of the performance data during UL TBF establishment, the engineers found that most UL TBF establishment failures occurred during the link establishment process of two-phase access on CCCH. The following table shows the performance data of the whole network during busy hours at night.



18


Numb er of L TBF Establ ishme nt Reque sts (on CCCH )

Numb er of ailed UL TBF Establ ishme nt Reque st s(Due to OnePhase Comp etition Failur e on CCCH )

Number Numb of failed er of UL TBF failed Establis UL hment TBF Reques Establ ts ishme (Resour nt ce Reque Reques sts t is not ( Due Receive to d on Other CCCH Factor During s on TwoCCCH Phase ) Access)

Start Time

Co mp leti on tim e

Num ber of UL TBF Esta blish ment Req uest s (UL Reso urce Avail able)

Num ber of UL TBF Esta blish ment Req uest s (on PAC CH)

Numb er of ailed UL TBF Establ ishme nt Reque sts (Time out on PACC H)

Numb er of failed UL TBF Establ ishme nt Reque sts (Due to Other Factor s on PACC H)

200811-16

200 81117

1997 6048

6269 943

3

21160

115595 45

6

1043693

115955

200811-17

200 81118

19671 183

6031 754

8

19028

111527 43

118

934292

112291

The above table indicates that UL TBF establishment failures caused by resource request not received on the CCCH during two-phase access account for 87%-89%. Therefore, it is crucial to reduce this kind of failures. In order to reduce this kind of failures and simplify TBF establishment process to speed up TBF establishment, the engineers decided to enable one-phase access. That requires modifying the parameter of TwoPhaseAccess from Yes to No.

5.1.3

Effect of Parameter Modification After the value of TwoPhaseAccess was modified, the UL TBF establishment success rate was greatly improved. The following table shows the performance data of the whole network during busy hours at night after the parameter was modified.



19


Start Time

Compl etion time

Numb er of UL TBF Establ ishme nt Reque sts (UL Resou rce Availa ble)

200811-18

200811-19

249721 28

631183 6

10879732

224337 59

19072081

98.44%

200811-19

200811-20

243067 75

603344 9

10678961

220544 48

18317672

98.42%

200811-20

200811-21

241556 75

595766 7

10635179

220661 26

18065659

98.49%

Numb er of UL TBF Establ ishme nt Reque sts (on PACC H)

Number of UL TBF Establish ment Request (on CCCH)

Numb er of UL Signal ing TBF Establ ishme nt Times

Number of Number of UL Data UL TBF TBF Establishme Establish nt Success ments Rate (%) Times

According to the above table, t he UL TBF establishment rate of the whole network after the network modification is about 98.4%, which is a satisfactory result. Further observation of the performance data during UL TBF establishment showed that the number of UL TBF establishment failures caused by resource request not received on the CCCH during two-phase access decreased dramatically. The following table shows the performance data of the whole network during busy hours at night.


Num ber of faile d UL TBF Esta blish ment Req uest s (Tim eout on PAC CH)


Numb er of UL TBF Establ ishme nt Reque sts (on CCCH )

Numb er of failed UL TBF Establ ishme nt Reque sts (Due to OnePhase Comp etition Failur e on CCCH )


Start Time

Com pleti on time


200811-18

200811-19

2497 2128

63118 36

1

29046

108797 32

275

217515

47614

200811-19

200811-20

2430 6775

6033 449

2

27315

106789 61

626

218043

44392

2008-

2008-

2415

5957

82

28368

106351

783

218027

45207



20


Start Time

Com pleti on time


11-20

11-21

5675


667

Num ber of faile d UL TBF Esta blish ment Req uest s (Tim eout on PAC CH)


Numb er of UL TBF Establ ishme nt Reque sts (on CCCH )

Numb er of failed UL TBF Establ ishme nt Reque sts (Due to OnePhase Comp etition Failur e on CCCH )


79

According to the above table, the number of UL TBF establishment failures caused by resource request not received on the CCCH during two-phase access decreased from one million to around 210,000. Therefore, the parameter modification is quite effective.

5.1.4

Verification of the Parameter Modification Effect Test site 1: an exhibition center The screenshot of the test result is as follows.

Test result: The engineers conducted a WAP login test for the BTS at the exhibition center and found that the WAP service of the BTS was normal and the one-phase access function was also normal.



21


Test site 2: Funan Primary School The screenshot of the test result is as follows.

Test result: The engineers conducted a WAP login test for the BTS of Funan Primary School and found that the WAP service of this BTS was normal and the one-phase access function was normal too. Test site 3: Xinshikong Building The screenshot of the test result is as follows.

Test result: The engineers conducted a WAP login test for the BTS of the Xinshikong Building and found that the WAP service of this BTS was normal and the one-phase access function was normal too.

5.1.5

Change of Other Indicators After the Parameter Modification The modification of the parameter of TwoPhaseAccess only affects TBF establishment. Therefore, we only need to look at the indicators that are relevant to the TBF



22


establishment process. The following table shows the relevant performance data of the whole network during busy hours at night.

Numb er of PRS UL TBF Data Establ ishme nt Times

Numb Number er of of GPRS GPRS UL UL TBF TBF Abnorm Abnor al mal Release Relea s Due ses to Due to N3101 N3103 Overflo Overfl w ow

Numb er of GPRS UL TBF Relea ses Due to Force d Relea se

Number of Number GPRS of UL TBF GPRS Release UL TBF s Due Release to s Due Instanc to Inner e Abnorm Suspen al sion

Start Time

Compl etion time

Numb er of PRS UL TBF Signal ing Establ ishme nt Times

200811-16

200811-17

107399 71

258468 88

860180

325270

140031

1658

660098

200811-17

200811-18

115106 53

242206 14

792474

318434

112116

1421

619080

200811-18

200811-19

224337 59

190720 81

526025

421180

101486

970

621291

200811-19

200811-20

220544 48

183176 72

506556

402282

95179

816

594196

200811-20

200811-21

220661 26

180656 59

500110

406515

102641

839

582291

The above table indicates that after the parameter modification, the number of successful GPRS UL signaling TBF establishments improved a lot. This is mainly because the number of establishment failures reduced after the parameter modification. However, because of various factors, the number of successful GPRS UL data TBF establishments did not change much. Therefore, parameter modification improved the stability of the network performance, and the influence on other indicators was very little.

5.2

A Case of TBF Establishment Success Rate Improvement for a Network in China (2)

5.2.1

Background Description In one ZTE network, the GPRS UL TBF establishment success rate was only 92% on average, which was very low.

5.2.2

Cause Analysis and Procedure Through the analysis of UL TBF establishment success rate, there were a lot of failures that were caused by resource request not received during the two-phase access.



23


Time Granula rity

GERAN SubNetwork

UL TBF Establis hment Succes s Rate

Number of GPRS UL TBF Establishmen t Failures (Due to OnePhase Competition Failure)

Number of GPRS UL TBF Establishmen t Failures (CCCH/PCCC H) (Resource Request is not Received During TwoPhase Access)

Number of GPRS UL TBF establishm ent Failures (CCCH/PC CCH) (Other Reasons)

1.18 22:0023:00

Tongliang, Tongnan iBSC (19)

95.69%

0

6476

495

1.19 22:0023:00

Tongliang, Tongnan iBSC(19)

95.64%

0

6232

543

1.20 22:0023:00

Tongliang, Tongnan iBSC(19)

95.76%

0

6855

587

At present, the parameter of TwoPhaseAccess is Yes by default. Because the RLC layer adopts acknowledged mode, generally data TBF is established in two-phase access mode (whether the MS access bit is not more than 8 blocks or more than 8 blocks). That is, whether EGPRS PACKET CHANNEL REQ is supported, the MS will adopt two-phase access mode. In this mode, the network must receive PACKET RESORCE REQUEST sent from the MS on a single block. There is only one chance to receive this message. If the UL BLER is high or data is lost on the Abis interface, the BSC is very likely to fail to receive the single block. When the value of TwoPhaseAccess is set to No, the MS is allowed to adopt one-phase access. In this case, which access mode is to be used is determined by whether the MS supports EGPRS PACKET CHANNEL REQ. Usually, the MS adopts one-phase access for signaling TBF. Therefore, when one-phase access is allowed, TBF establishment success rate can be improved. Solution: The field engineer disabled the compulsory two-phase access mode. That is, to set the TwoPhaseAccess parameter on module 3 and module 4 to No. The following table shows the data comparison before and after the modification.



24


Time

Bef ore Mod ifica tion

Afte r Mod ifica tion

GERAN Subnetwork

UL TBF Establish ment Success Rate

Number of GPRS UL TBF Establish ment Failures (Due to One-Phase Competitio n Failure)

Number of GPRS UL TBF Establishme nt Failures (CCCH/PCC CH) (Resource Request Is not Received During TwoPhase Access)

Number of GPRS UL TBF establishm ent Failures (CCCH/PC CCH) (Other Reasons)

1.18

Tongliang,Tongn an iBSC(19)

95.69%

0

6476

495

1.19


95.64%

0

6232

543

1.20


95.76%

0

6855

587

1.21


96.68%

0

3941

451

1.22


96.51%

1

4474

499

1.23


96.18%

0

4695

530

5.3

A Case of TBF Establishment Success Rate Improvement for an Overseas Network

5.3.1

Background Description The UL TBF establishment success rate of one ZTE’s overseas network is just around 90% on average during 24 hours, which is far from the operator’s requirement (96%).

5.3.2

Cause Analysis and Procedure The analysis of UL TBF establishment success rate showed that a lot of UL TBF establishment failures were caused by resource request not received on the CCCH during two-phase access. At present, the parameter of TwoPhaseAccess is set to Yes by default. Because the RLC layer adopts acknowledged mode, generally data TBF is established in two-phase access mode (whether the MS access bit is not more than 8 blocks or more than 8 blocks). That is, whether EGPRS PACKET CHANNEL REQ is supported, the MS will adopt the two-phase access mode. In this mode, the network must receive PACKET RESORCE REQUEST sent from the MS on a single block. There is only one chance to receive it. If the UL BLER is high or data is lost on the Abis



25


interface, the BSC is very likely to fail to receive the single block. When the value of TwoPhaseAccess is set to No, the MS is allowed to adopt one-phase access mode. In this case, which access mode is to be used is determined by whether the MS supports EGPRS PACKET CHANNEL REQ. Usually, the MS adopts one-phase access for signaling TBF. Therefore, when one-phase access is allowed, TBF establishment success rate can be improved. Based on this analysis, the field engineer adjusted BSC1 of the current network on March 22 to allow one-phase access. The following table shows the data collected before and after the modification.

Object

Time

ADDISGPRS UL TBF Establish ment Success Rate

ADDIS-OnePhase Competition Failure

ADDISResource Request Is not Received During TwoPhase Access

ADDISNumber of GPRS UL TBF Establish ment Failures (CCCH/PC CCH) (Other Reasons)

BSC1

2010-03-13

92.07

2

65680

7075

BSC1

2010-03-14

93.29

0

52695

7394

BSC1

2010-03-15

89.45

0

85692

6645

BSC1

2010-03-16

89.39

2

99289

8884

BSC1

2010-03-17

90.98

0

84783

6445

BSC1

2010-03-18

90.46

0

84738

5820

BSC1

2010-03-19

90.74

0

107911

6098

BSC1

2010-03-20

88.88

3

134056

6472

BSC1

2010-03-21

92.35

1

75041

6705

BSC1

2010-03-22

93.2

2

37118

2849

BSC1

2010-03-23

96.8

15

3169

202

BSC1

2010-03-24

96.36

4

3226

212

BSC1

2010-03-25

96.98

4

3365

242

BSC1

2010-03-26

96.69

1

3874

369

BSC1

2010-03-27

97.24

1

3769

421

BSC1

2010-03-28

97.55

7

2467

418

The operator’s requirement of TBF establishment success rate for this network is 96%. After the two-phase access mode is changed to one-phase access mode, UL TBF establishment success rate is greatly increased to meet this requirement.



26




27


6

Related Measurement and Counters

6.1

PS Basic Measurement Counter ID

Counter Name

C900040108

Number of GPRS DL TBF establishment requests

C900040109

Number of GPRS DL TBF establishment failures due to database allocation failure

C900040110

Number of GPRS DL TBF resource reallocation requests due to UL TBF establishment

C900040111

Number of GPRS DL TBF resource reallocation failures due to UL TBF establishment

C900040112

Number of GPRS DL TBF resource reallocation requests due to LLC transmission

C900040113

Number of GPRS DL TBF resource reallocation failures due to LLC transmission

C900040114

Number of EGRPS DL TBF establishment requests

C900040115

Number of EGPRS DL TBF establishment failures due to database allocation failure

C900040116

Number of EGPRS DL TBF resource reallocation requests due to UL TBF establishment

C900040117

Number of GPRS DL TBF resource reallocation failures due to UL TBF establishment

C900040118

Number of EGPRS DL TBF resource reallocation requests due to LLC transmission

C900040119

Number of EGPRS DL TBF resource reallocation failures due to LLC transmission

C900040120

Number of GPRS UL TBF establishment requests

C900040121

Number of GPRS UL TBF establishment failure due to database allocation failure

C900040122

Number of GPRS UL TBF resource reallocation requests due to DL TBF establishment

C900040123

Number of GPRS UL TBF resource reallocation failures due to DL TBF establishment

C900040124

Number of GPRS UL TBF resource reallocation requests due to LLC transmission

C900040125

Number of GPRS UL TBF resource reallocation failure due to LLC transmission

C900040126

Number of GPRS UL TBF resource reallocation requests for resource request

C900040127

Number of GPRS UL TBF resource reallocation failure due to resource request



28


Counter ID

Counter Name

C900040128

Number of EGPRS UL TBF establishment requests

C900040129

Number of EGPRS UL TBF establishment failure due to database allocation failure

C900040130

Number of EGPRS UL TBF resource reallocation requests due to DL TBF establishment

C900040131

Number of EGPRS UL TBF resource reallocation failures due to DL TBF establishment

C900040132

Number of EGPRS UL TBF resource reallocation requests due to LLC transmission

C900040133

Number of EGPRS UL TBF resource reallocation failure due to LLC transmission

C900040134

Number of EGPRS UL TBF resource reallocation requests for resource request

C900040135

C901200067:Number of EGPRS UL TBF resource reallocation failure due to resource request

C900040136

Average number of available PDCH

C900040137

Number of forcing conversation dynamic from PDCH to TCH by BSC

C900040138

Number of PDCH used by TBF in each cell

C900040139

Maximum number of used PDCH

C900040140

Average number of used PDCH

C900040141

Number of GPRS DL TBF establishment requests before T1392 expiration

C900040142

Number of GPRS DL TBF establishment using existing DL TBF resource

C900040143

Number of GPRS DL TBF establishment requests on PACCH

C900040144

Number of GPRS CCCH/PCCCH

C900040145

Number of EGPRS DL TBF establishment requests before T1392 expiration

C900040146

Number of EGPRS DL TBF establishment using DL TBF resource

C900040147

Number of EGPRS DL TBF establishment requests on PACCH

C900040148

Number of EGPRS CCCH/PCCCH

C900040149

Number of GPRS DL signaling TBF release due to N3105 overflow

C900040150

Number of GPRS DL signaling TBF release due to other reasons.

C900040151

Number of GPRS DL signaling TBF forcing release by control plane

C900040152

Number of GPRS DL signaling TBF release due to instance suspension.

C900040153

Number of GPRS DL signaling TBF release due to inner abnormal.

C900040154

Number of EGPRS DL signaling TBF release due to N3105 overflow

C900040155

Number of EGPRS DL signaling TBF release due to other reasons


DL

DL

TBF

TBF

establishment

establishment


requests

requests

29

on

on


Counter ID

6.2

Counter Name

C900040156

Number of EGPRS DL signaling TBF forcing release by control plane

C900040157

Number of EGPRS DL signaling TBF release due to instance suspension

C900040158

Number of EGPRS DL signaling TBF release due to inner abnormal

C900040159

Number of GPRS UL TBF establishments in release state

C900040160

Number of GPRS UL TBF establishment with existing UL resource

C900040161

Number of GPRS UL TBF establishment requests on PACCH

C900040162

Number of EGPRS UL data TBF establishments on PACCH

C900040163

Number of EGPRS UL TBF establishments in release state

C900040164

Number of EGPRS UL TBF establishment with existing UL resource

C900040165

Number of EGPRS UL TBF establishment requests on PACCH

C900040166


C900040167

Number of dynamic PDCH preemption for CS service

C900040168


C900040169

Number of GPRS DL data TBF establishments on PACCH

C900040170

Number of EGPRS DL data TBF establishments on PACCH

C900040171

Number of GPRS UL data TBF establishments on PACCH

UL

UL

TBF

TBF

establishment

establishment

requests

on

requests

on

Usage of PS Resources Counter ID

Counter Name

C901040001

Maximum number of Abis TS used

C901040002

Average number of Abis TS used

C901040004

Average number of unavailable PDCH

C901040005

Number of available static PDCH

C901040006

Number of unavailable static PDCH

C901040009

Time of DL PDCH congestion

C901040010

Time of UL PDCH congestion

C901040011

Time of PDCH used at least one UL/DL TBF exists

C901040012

Time of PDCH used at least one UL/DL EGPRS TBF exists

C901040013

Time of PDCH used at least one DL TBF exists

C901040014

Time of PDCH used at least one UL TBF exists



30


6.3

UL/DL GPRS/EDGE Data TBF Counter ID

Counter Name

C901190002

Number of GPRS UL TBF establishment requests on CCCH

C901190003

Number of EGPRS UL TBF establishment requests on CCCH

C901190004

Number of GPRS UL TBF establishment requests on PCCCH

C901190005

Number of EGPRS UL TBF establishment requests on PCCCH

C901190006


C901190015

Number of GPRS UL TBF establishment failure due to Abis congestion

C901190016

Number of GPRS UL TBF establishment failure due to RR congestion

C901190017

Number of GPRS UL TBF establishment failure due to UPPB-DSP channel congestion

C901190018

Number of GPRS UL TBF establishment channel allocate success

C901200002

Number of GPRS UL TBF establishment requests on CCCH

C901200003

Number of EGPRS UL TBF establishment requests on CCCH

C901200004

Number of GPRS UL TBF establishment requests on PCCCH

C901200005

Number of EGPRS UL TBF establishment requests on PCCCH

C901200006


C901200015

Number of EGPRS UL TBF establishment failure due to Abis congestion

C901200016

Number of EGPRS UL TBF establishment failure due to RR congestion

C901200017

Number of EGPRS UL TBF establishment failure due to UPPBDSP channel congestion

C901200018

Number of EGPRS UL TBF establishment channel allocate success

C901170002


C901170003

Number of GPRS DL TBF establishment requests on PACCH

C901170004

Number of GPRS DL TBF establishment failure due to CPU

C901170005

Number of GPRS DL TBF establishment failure due to not support DTM

C901170007

Number of GPRS DL TBF establishment failure due to Abis congestion

C901170008

Number of GPRS DL TBF establishment failure due to RR congestion

C901170009

Number of GPRS DL TBF establishment failure due to UPPB-DSP channel congestion

C901170010

Number of GPRS DL TBF establishment channel allocate success

C901180002


C901180003

Number of EGPRS DL TBF establishment requests on PACCH

C901180004

Number of EGPRS DL TBF establishment failures due to CPU


DL

TBF

DL TBF

establishment

establishment


requests

requests

31

on

on

TBF Establishment Success Rate(ZTE)

Recommend Documents