Intra-frequency measurements Reporting event 1A: A Primary CPICH enters the reporting range Reporting event 1B: A primary CPICH leaves the reporting range Reporting event 1C: A non-active primary CPICH becomes better than an active primary CPICH Reporting event 1D: Change of best cell Reporting event 1E: A Primary CPICH becomes better than an absolute threshold Reporting event 1F: A Primary CPICH becomes worse than an absolute threshold Intra-frequency reporting events for TDD Reporting event 1G: Change of best cell (TDD) Reporting event 1H: Timeslot ISCP below a certain threshold (TDD) Reporting event 1I: Timeslot ISCP above a certain threshold (TDD) Inter-frequency measurements Event 2a: Change of best frequency. Event 2b: The estimated quality of the currently used frequency is below a certain threshold and the estimated quality of a non-used frequency is above a certain threshold Event 2c: The estimated quality of a non-used frequency is above a certain threshold Event 2d: The estimated quality of the currently used frequency is below a certain threshold Event 2e: The estimated quality of a non-used frequency is below a certain threshold Event 2f: The estimated quality of the currently used frequency is above a certain threshold Inter-RAT measurements Event 3a: The estimated quality of the currently used UTRAN frequency is below a certain threshold and the estimated quality of the other system is above a certain threshold Event 3b: The estimated quality of other system is below a certain threshold Event 3c: The estimated quality of other system is above a certain threshold Event 3d: Change of best cell in other system UE internal measurements Reporting event 6A: The UE Tx power becomes larger than an absolute threshold Reporting event 6B: The UE Tx power becomes less than an absolute threshold Reporting event 6C: The UE Tx power reaches its minimum value Reporting event 6D: The UE Tx power reaches its maximum value Reporting event 6E: The UE RSSI reaches the UE's dynamic receiver range Reporting event 6F: The UE Rx-Tx time difference for a RL included in the active set becomes larger than an absolute threshold Reporting event 6G: The UE Rx-Tx time difference for a RL included in the active set becomes less than an absolute threshold

SDCCH Drop: SDCCH Drop Reasons and Solutions

1)Low Signal Strength on Down or Uplink The reason for poor coverage could be too few sites, wrong output power, shadowing, no indoor coverage or network equipment failure. Action: Check coverage plots.Check output power. Perform drive tests. Check BTS error log Solution: Add new sites. Increase output power. Repair faulty equipment.

2)Poor Quality on Down or Uplink Action: Check C/I and C/A plots. Check frequency plan. Perform drive tests. Solution: Change frequency. Use available radio features.

3)Too High Timing Advance Action: Check if the cell parameter TALIM is < style="font-weight: bold;">Solution: Set TALIM to a value close to 63. Tilt antenna/reduce antenna height/output power, etc. for cochannel cells.

4)Mobile Error Some old mobiles may cause dropped calls if certain radio network features are used. Another reason is that the MS is damaged and not working properly. Action: Check MS fleet. Solution: Inform operator.

5)Subscriber Behavior Poorly educated subscribers could use their handsets incorrectly by not raising antennas, choosing illadvised locations to attempt calls, etc. Action: Check customer complaints and their MS.

6)Battery Flaw When a subscriber runs out of battery during a conversation, the call will be registered as dropped call due to low signal strength or others. Action: Check if MS power regulation is used. Check if DTX uplink is used.

7)Congestion on TCH The SDCCH is dropped when congestion on TCH. Action: Check TCH congestion Solution: Increase capacity on TCH or using features like Assignment to another cell, Cell Load Sharing, HCS, Dynamic Half-Rate Allocation and FR-HR Mode Adaptation etc

Dropped Call Analysis (TCH Drop Analysis): Drop Analysis.

1. Radio Link Time-Out Every time a SACCH message can not be decoded the radio link time-out counter is decreased by 1. If the message can be decoded the counter is incremented by 2. However, the value can not exceed the initial value. The initial value is set by the parameter RLINKT for radio link time-out in the mobile station and by RLINKUP for timeout in the BSC. If the mobile moves out of coverage and no measurement reports are received in the BSC, there will be a radio link time-out and the message Channel Release (cause: abnormal release, unspecified) is sent to the mobile station and the SACCH is deactivated in the BTS. A Clear Request message is sent to the MSC. To be sure that the mobile has stopped transmitting, the BSC now waits RLINKT SACCH periods before the timeslot is released and a new call can be established on the channel. 2. Layer 2 Time-Out If the BTS never get an acknowledge on a Layer 2 message after the time T200XN200, the BTS will send Error Indication (cause: T200 expired) to the BSC, which will send Channel Release (cause: abnormal release, timer expired) to the mobile station and a Clear Request to the MSC. The SACCH is deactivated and the BSC waits RLINKT SACCH periods before the timeslot is released and a new call can use the channel. This is only valid if the call is in steady state, i.e. not during handover or assignment.

3. Release Indication When the BTS received a layer 2 DISC frame from the mobile it replies with a Layer 2 UA frame to the mobile station and a Release Indication to the BSC. The system does only react on Release Indication if it is received during a normal disconnection situation. If such a message is received unexpectedly this will usually cause radio

link time-out or timer T200 expiration as the mobile station stops the transmitting of measurement reports. It is also possible that the release will be normal depending on when the Release Indication is received. 4. MSC Time-Out Normal Release: If the MSC never received a response on a message (e.g. Identity Request) and there is no radio link time-out or layer 2 time-out, the MSC will send a Clear Command to the BSC. The time-out is depending on the message. When receiving Clear Command, the BSC will send a Channel Release (cause: normal release) and then deactivates the SACCH. Reject (only SDCCH): If the MSC never receives a response on the first message after Establish Indication, the MSC will send a reject message. If the connection was a Location Update it will be a Location Update Reject (cause: network failure) and if the connection was a mobile originating call (CM Service Request) a CM Service Reject (cause: network failure) will be sent. The MSC will then send a Clear Command to the BSC and the call is cleared by Channel Release (cause: normal release). 5. Assignment to TCH Before sending an Assignment Command from the BSC at TCH assignment, the following two criterion have to be fulfilled: a. There must be a TCH channel available, i.e. no congestion b. The locating algorithm must have received at least one valid measurement report. If either of the criterion is not fulfilled, Assignment Command will not be sent and a Channel Release (cause: abnormal release, unspecified) will be sent to the mobile station and a Clear Request to the MSC. TCH Drop reason (1) The classification of TCH Drop Reasons are arranged in the order of priority: 1.Excessive Timing Advance 2.Low Signal Strength 3.Bad Quality 4.Sudden Loss of Connection 5.Other Reasons Excessive Timing Advance The TCH Drop counters due to Excessive Timing Advance will pegged when the during the time of disconnection, the last Timing Advance value recorded was higher than the TALIM Parameter. This drop reason is commonly apparent to isolated or island sites with a wide coverage area. Action: Check if the cell parameter TALIM is < "63" Solution: Set TALIM to a value close to 63. Tilt antenna/reduce antenna height/output power, etc. for co-channel cells.

TCH Drop Reasons (2) Low Signal Strength on Down or Uplink or Both Links The drops counters due to Low Signal Strength will be pegged when the Signal Strength during the last Measurement Report before the call dropped is below the LOWSSDL and/or LOWSSUL Thresholds. LOWSSDL and LOWSSUL are BSC Exchange Property parameters which is used only for statistics purposes and does not affect the behavior of calls. If both UL and DL Signal Strength are below the thresholds, only Drop due to Low SS BL will pegged. Normally a call is dropped at the border of large rural cell with insufficient coverage. Bad tunnel coverage cause many dropped calls as well as so called coverage holes. Bad indoor coverage will result in dropped calls. Building shadowing could be another reason. Action: Check coverage plots. Check output power. Check power balance and link budget. Check if Omni site. Check antenna configuration & type. Check antenna installation. Perform drive tests & site survey. Check TRX/TS with high CONERRCNT. Solution: Add a repeater to increase coverage in for example a tunnel. Change to a better antenna (with higher gain) for the base station. Add a new base station if there are large coverage holes. Block/Deblock TRX

TCH Drop Reasons (3) Poor Quality on Down or Uplink or Both Links The drops counters due to Bad Quality will be pegged when the Signal Strength during the last Measurement Report before the call dropped is above the BADQDL and/or BADQUL Thresholds. BADQDL and BADQUL (expressed in DTQU) are BSC Exchange Property parameters which is used only for statistics purposes and does not affect the behavior of calls. If both UL and DL Quality are above the thresholds, only Drop due to BAD Quality BL will pegged. Problem on Bad Quality is usually associated with Co-channel Interference on BCCH or TCH. Faulty MAIO assignment can cause frequency collisions on co-sited cells especially on 1x1 Reuse. External interference is also one possible cause of problem on quality. Action: Check C/I and C/A plots. Check Frequency Plan (Co-BCCH or Co-BSIC Problem). Check MAIO, HOP, HSN parameters. Check FHOP if correctly configured (BB or SY). Check for External Interference. Perform drive tests. Solution: Change BCCH frequency. Change BSIC. Change MAIO, HOP, HSN. Change FHOP. Record RIR or on-site Frequency Scanning to identify source of interference. Use available radio features. TCH Drop Reasons (4) Sudden Loss of Connection Drops due to Sudden Loss are drops that have not been registered as low signal strength, excessive timing advance, bad quality or hardware (other) reasons, and the locating procedure indicates missing measurement results from the MS. There are some common scenarios that could lead to Sudden Loss of connections such as very sudden and severe drops in signal strength, such as when subscribers enter into buildings, elevators, parking garages, etc., very sudden and severe occurrence of interference, MS runs out of battery during conversation, Handover Lost, BTS HW faults, Synchronization or A-bis link fault (transmission faults), and MS Faults. Action: Check BTS Error Logs, Alarms and Fault Codes. Check CONERRCNT per TRX and TS. Check Transmission Link (A-bis). Check for DIP Slips. Check LAPD Congestion. Correlate Handover Lost to Drops due to Sudden Loss Solution: Fix Hardware Faults and Alarms. Reset TRX with high CONERRCNT. Ensure that Synchronization and A-bis Link are stable. Change RBLT with high DIP Slips. Change CONFACT or increase Transmission Capacity Investigate HO Lost Problem TCH Drop Reasons (5) TCH Drops due to Other Reasons TCH drops due to Other Reasons are computed by subtracting the sum of drops due to Excessive TA, Low SS, Bad Quality and Sudden Loss from the Total TCH Drop Counts. Drops due to Other Reasons are generally associated with hardware problems, transmission link problems on A-bis, Ater or Ainterfaces, and sometimes Handover Lost. Action: Check BTS Error Logs. Check Alarms and Fault Codes. Check CONERRCNT per TRX and TS. Check Transmission Link (A-bis).

Check for DIP Slips. Correlate Handover Lost to Drops due to Other Reasons Solution: Fix Hardware Faults and Alarms. Reset TRX with high CONERRCNT. Ensure that Synchronization and A-bis Link are stable. Change RBLT with high DIP Slips. Investigate HO Lost Problem

Problem reason of drop in SDCCH Low Signal Strength on Down or Uplink The reason for poor coverage could be too few sites, wrong output power, shadowing, no indoor coverage or network equipment failure. Action: Check coverage plots.Check output power. Perform drive tests. Check BTS error log Solution: Add new sites. Increase output power. Repair faulty equipment. Poor Quality on Down or Uplink Action: Check C/I and C/A plots. Check frequency plan. Perform drive tests. Solution: Change frequency. Use available radio features. Too High Timing Advance Action: Check if the cell parameter TALIM is < style="font-weight: bold;">Solution: Set TALIM to a value close to 63. Tilt antenna/reduce antenna height/output power, etc. for cochannel cells. Mobile Error Some old mobiles may cause dropped calls if certain radio network features are used. Another reason is that the MS is damaged and not working properly. Action: Check MS fleet. Solution: Inform operator. Subscriber Behavior Poorly educated subscribers could use their handsets incorrectly by not raising antennas, choosing illadvised locations to attempt calls, etc. Action: Check customer complaints and their MS. Battery Flaw When a subscriber runs out of battery during a conversation, the call will be registered as dropped call due to low signal strength or others. Action: Check if MS power regulation is used. Check if DTX uplink is used. Congestion on TCH The SDCCH is dropped when congestion on TCH. Action: Check TCH congestion Solution: Increase capacity on TCH or using features like Assignment to another cell, Cell Load Sharing, HCS, Dynamic Half-Rate Allocation and FR-HR Mode Adaptation etc

What are the optimization tools you use? Drive test, analysis, others? 2. Are System Information Blocks (SIB) transmitted all the time? No, system information block is multiplexed with synchronization channel. Synchronization channel occupies the first time slot (TS) and SIB occupies the other 9 time slots. 3. How does UE camp (synchronize) to a NodeB?

1. UE uses the primary synchronization channel (P-SCH) for slot alignment (TS synchronization). 2. After aligning to NodeB time slot, UE then uses secondary synchronization channel (S-SCH) to obtain frame synchronization and scrambling code group identification. 3. UE then uses scrambling code ID to obtain CPICH, thus camping to a NodeB. 4. What could be the cause of soft handover failure?

UE issue. Resource unavailable at target NodeB. Inadequate SHO threshold defined. o Etc.

5. What are the three sets in handover? The 3 sets in handover are:

Active set the list of cells which are in soft handover with UE. Monitored set the list of cells not in active set but RNC has told UE to monitor. o Detected set list of cells detected by the UE but not configured in the neighbor list.

6. What are the major differences between GSM and UMTS handover decision? GSM:

Time-based mobile measures of RxLev and RxQual mobile sends measurement report every SACH period (480ms). o BSC instructs mobile to handover based on these reports.

UMTS:

Event-triggered reporting UE sends a measurement report only on certain event triggers. o UE plays more part in the handover decision.

7. What are the events 1a, 1b, 1c, etc.?

e1a a Primary CPICH enters the reporting range, i.e. add a cell to active set. e1b a primary CPICH leaves the reporting range, i.e. removed a cell from active set. e1c a non-active primary CPICH becomes better than an active primary CPICH, i.e. replace a cell. e1d: change of best cell. e1e: a Primary CPICH becomes better than an absolute threshold. o e1f: a Primary CPICH becomes worse than an absolute threshold.

8. What are event 2a-2d and 3a-3d? Events 2a-2d are for inter-frequency handover measurements and events 3a-3d are for IRAT handover measurements.

e3a: the UMTS cell quality has moved below a threshold and a GSM cell quality had moved above a threshold. e3b: the GSM cell quality has moved below a threshold. e3c: the GSM cell quality has moved above a threshold. e3d: there was a change in the order of best GSM cell list.

9. What may happen when theres a missing neighbor or an incorrect neighbor?

Access failure and handover failure: may attempt to access to a wrong scrambling code.

Dropped call: UE not aware of a strong scrambling code, strong interference. Poor data throughput. Poor voice quality. Etc.

10. What can we try to improve when access failure is high? When access failure is high we can try the following to improve RACH performance:

Increase maximum UE transmit power allowed: Max_allowed_UL_TX_Power. Increase power quickly: power_Offset_P0. Increase number of preambles sent in a given preamble cycle: preamble_Retrans_Max. Increase the number of preamble cycles: max_Preamble_Cycle. Increase number of RRC Connection Request retries: N300.

11. What are the conditions you typically set to trigger IRAT handover? RSCP and Ec/Io are used to trigger IRAT handover:

RSCP -100dBm. Ec/Io -16dBm.

12. What are the typical KPIs you use to measure a network and what criteria?

Access failure rate ( 2%). Call setup time (CS: over 95% of the time < 6-second for mobile-to-PSTN, 9-second for mobile-mobile. PS: over 95% of the time < 5-second). Dropped call rate ( 2%). BLER: over 95% of the blocks 2%. Average DL/UL throughput for PSD: 210kbps for loaded, 240kbps for unloaded.

13. What is the typical UE transmit power? Varies most of the time below 0dBm. 14. Have your used Ericsson TEMS? If so:

Do you know how to create command sequence? What are the call sequences you typically have? CS long call, CS short call, PSD call, etc. What are the typical commands you have for CS and PS call? Do you regularly stop and restart a new log file? Why and when to stop and start a new file? How do you stop a log file? Stop command sequence first, wait and make sure all equipment are in idle mode before stop logging.

15. Did you work on neighbor prioritization? 16. What is the typical event sequence of IRAT Handover from 3G to 2G

Event 2d entering into compressed mode measurement of 2G candidates Event 3a Verification of 2G resources Handover from UTRAN Command from 3G RNC to UE

17. What are the possible causes for an IRAT Failure?

Missing 2G relations Non availability of 2G Resources Poor 2G Coverage Missing 3G Relations

18. What is Paging Success Ratio? What is the typical PSR that you have seen in a UMTS network?

PSR Paging Responses to the Paging Attempts About 90%

19. What are the possible causes for a lower PSR?

Non-continuous RF Coverage UE going in and out of coverage area frequently Very High Periodic Location Update Timer Keeping UEs in VLR long time after it moved out of coverage Lower Paging Channel Power Access Channel Parameter Issues Delayed Location Update when crossing the LA / CN Boundaries

20. What are the possible causes for a Drop Call on a UMTS network?

Poor Coverage (DL / UL) Pilot Pollution / Pilot Spillover Missing Neighbor SC Collisions Delayed Handovers No resource availability (Congestion) for Hand in Loss of Synchronization Fast Fading Delayed IRAT Triggers Hardware Issues External Interferencerd

21. A UE is served by 2 or 3 SC in AS. It is identifying a SC from 3 tier, Stronger and meets the criteria for Event1a or Event1c. But SHO did not happen because of missing neighbor relations? How do you optimize this issue?

Study the Pilot spillover from the 3 Tier SC and control its coverage Even after controlling the coverage, if the spillover is there, Add the neighbor.

rd

22. A UE is served by 2 SC in AS, a SC is coming in to Monitored Set and Event1a is triggered. But UE is not receiving Active Set Update from NodeB and the call drops. What could be possible causes for this drop?

Delayed Handover Loss of Synchronization Fast Fading Pilot Pollution / Spillover issues

23. What is Hard Handover in UMTS? When will it happen?

Hard Handover in UMTS is a break before make type Handover It can happen in the inter RNC boundaries where there is no Iur link.

24. What is the typical Call Setup Time for a 3G UE to 3G UE Call? What are the possible RF related causes for a delayed CST in this type of call?

6 to 9 seconds Multiple RRC Attempts (UE is on poor coverage need more than Access Attempt) Delayed Page Responses High Load on Paging and/or Access Channel Paging / Access Parameters

25. What is Soft Handover Overhead? What is the typical value in UMTS network?

Soft Handover Overhead is calculated in two ways. 1) Average Active Set Size Total Traffic / Primary Traffic. 2) Secondary / Total Traffic Typical Values are like 1.7 (Avg Active Set Size) or 35% (Secondary / Total )

26. What will happen to the Soft Handover Overhead when you apply OCNS on the network? And Why?

With OCNS, the interference (load) increases. This leads to reduction in Ec/Io of a Pilot, which reduces the pilot spillovers. Reduction in Pilot Spillover will reduce the Soft Handover Overhead.

27. What are the possible causes for an Access Failure in UMTS?

Missing Neighbors Poor Coverage Pilot Pollution / Spillover Poor Cell Reselection Core Network Issues Non availability of resources. Admission Control denies Hardware Issues Improper RACH Parameters External Interference

28. (FOR ERICSSON EXPERIENCED) What is RTWP? What is the significance of it?

Received Total Wide-band Power It gives the Total Uplink Power (Interference) level received at NodeB

29. (FOR ERICSSON EXPERIENCED) What is the System Reference Point at which all the Power Levels are measured in Ericsson NodeB?

System Ref Point for E/// NodeB is at the output of TMA (Between TMA and Antenna)

30. What are the typical values for reportingrange1a and reportingrange1b?

3 dB and 5 dB respectively.

31. What will be the impact when you change reportingrange1a from 3 to 4 dB and timetotrigger1a 100 to 320 ms, without changing any other parameters?

Reduction in number of Event1a Delayed Event1a trigger Reduction in Average Active Set Size Delay in Event1a could increase DL interference, which could lead to a drop call or increase in Average Power Per User (reduction in cell capacity)

32. What is Admission Control?

Admission Control is an algorithm which controls the Resource Allocation for a new call and additional resource allocation for an existing call. Incase, if a cell is heavily a loaded and enough resources in terms of power, codes or CEs are not available, admission control denies permission for the additional resource requirement.

33. What is Congestion Control?

Congestion Control monitors the dynamic utilization of specific cell resources and insures that overload conditions do not occur. If overload conditions do occur, Congestion Control will immediately restrict Admission Control from granting additional resources. In addition, Congestion Control will attempt to

resolve the congestion by either down switching, or terminating existing users. Once the congestion is corrected, the congestion resolution actions will cease, and Admission Control will be enabled.

RRC Connection Setup Failure Analysis The RRC connection setup failure can be detected by the UE signaling process andRNC single-user trace. RRC connection is established via the following steps: UEsends the RRC Connection Request message via RACH, RNC sends the RRCConnection Setup message via FACH, then UE establishes the downlink dedicatedchannel, synchronizes it, and now sends the RRC Connection Setup CMP messagevia the uplink dedicated channel. Reasons for RRC connection failures include: Errors or faults in uplink RACH,downlink FACH power distribution, cell parameter reselection, low downlink initialtransmitting power, uplink initial power control, congestion, abnormal equipment, etc,especially those of uplink RACH, downlink FACH power distribution, cell parameterreselection, and abnormal equipment. The RRC Connection Request message sent by UE is not received by RNC. IfEc/Io of the downlink CPICH is not too low (say, about -12dB), then normallyRACH fails due to incorrect power control estimation for the uplink open loop, orinsufficient power escalation of Preamble, or lower-than-required output power ofUE. RNC, after having received the RRC connection request sent by UE, sends theRRC connection Setup message but UE fails to receive it. Now check the CPICHEc/Io. If it is blow -12dB and there is no better cell in the monitored set, then itmust be failure of coverage, in which case, increase the FACH power. If there isa better cell in the monitored set, it might be caused by cell reselection, in whichcase, adjust the cell reselection parameters to speed up the reselection. After having received the RRC Connection Setup message, UE fails to send theSetup Complete message. If the downlink signal quality is normal, then it mightbe caused by the abnormal mobile phone, or the initial power of the downlinkdedicated channel is too low to enable synchronization. UE has sent the RRC Setup Complete message but RNC fails to receive it.Since the uplink initial power control will make the UE transmitting powerescalate, such failure is rare. If it does happen, increase as appropriate theConstant Value of the dedicated channel.

WCDMA Radio Network Optimization Guide

Table of Contents Chapter 1 Introduction................................................. ............................................................1 -1 Chapter 2 Radio Netw ork Optimization Work Flow ................................ .................... 2-1 2.1 W ork Flow.....................................................................................................................2 -1 2.2 Related Report....................... .......................................................................................2 -2 2.2.1 Output Report and Table.....................................................................................2 -2 2.2.2 Process Table............................ .........................................................................2-2 Chapter 3 Project Preparation and Startup................................... ..........................................3 -1 3.1 Understanding of the Existing Network..........................................................................3 1 3.2 Optimization Team Establishment and Cluster Division.................................................3-1 3.2.1 RNO Team.........................................................................................................3 -1 3.2.2 Cluster Division...................................................................................................3 -2 3.3 Optimization Tools and Software................................................................. ..................32 3.3.1 Optimization Tool............................... .................................................................3-2 3.3.2 Optimization Software.........................................................................................3 -3 Chapter 4 Single Site Verification............................................................................................4 -1 4.1 Purpose........................................................................................................................4 -1 4.2 Preparation................................................................................................................... 4-1 4.2.1 Alarm Check................................. ......................................................................4 -1 4.2.2 Cell State Check.................................................................................................4 -1 4.2.3 Radio Parameters Check....................................................................................4 -1 4.3 Site Verification.............................................................................................................4 -2 Chapter 5 RF Optimization................ .......................................................................................5 -1 5.1 Purpose........................................................................................................................5 -1 5.2 Drive Survey.............................. ...................................................................................5-1

5.2.1 Preparation.........................................................................................................5 -1 5.2.2 Drive Test.............................. .............................................................................5-2 5.3 Data Analysis.......................................... ......................................................................5-2 5.4 Adjustment Recommendation and Implementation........................................................5-2 Chapter 6 Service Optimization........................................................... ....................................6-1 6.1 Purpose........................................................................................................................6 -1 6.2 Drive Survey.............................. ...................................................................................6-1 6.2.1 Preparation.........................................................................................................6 -1 6.2.2 Drive Test.............................. .............................................................................6-2 6.3 Data Analysis.......................................... ......................................................................6-2 6.4 Adjustment Recommendation and Implementation........................................................6-2 6.5 Test for Special Areas (CQT)........................................................................................6 3 6.6 Statistics Analysis.................................. .......................................................................6 -3 6.6.1 Preparation.........................................................................................................6 -3 6.6.2 Analysis..............................................................................................................6 -3 6.6.3 Adjustment Recommendation and Implementation..............................................6-3 6.7 Summary................................... ...................................................................................6 -3 Chapter 7 Output Report..........................................................................................................7 -1 7.1 Optimization Output Report and Table...........................................................................7 1 Chapter 8 Category of Optimization Problems and Case Analysis........................................8-1 8.1 Coverage Optimization................................. .................................................................8 -1 8.1.1 Classifying and Defining Coverage Problem........................................................8-1 8.1.2 Coverage Problem Analysis................................. ...............................................8-5 8.2 Call Drop Optimization...................................................... ...........................................816 8.2.1 Definition of Call Drop........................................................................................8 -16

8.2.2 Basic Flow for Analyzing Call Drop.....................................................................8-18 WCDMA Radio Network Optimization Guide Table of ContentsHuawei Technologies Proprietary3 8.2.3 Analysis for Usual Call Drop Reasons................................................................8-19 8.2.4 Parameter Configuration Related to Call Drop....................................................8-31 8.3 Access Optimization................... ..................................................................................8 -37 8.3.1 Classified Definitions of Access Failure..............................................................8-37 8.3.2 RRC Connection Setup Failure Analysis............................................................8-38 8.3.3 Paging Failure Analysis......................................................................................8-39 8.3.4 Access-related Parameter Configuration............................................................8-40 8.4 WCDMA Network Uplink Interference and Optimization................................................8-45 8.4.1 Cause Analysis of the Uplink Main Diversity RTWP Variance.............................8-46 8.4.2 Cause Analysis of the RTWP Abnormality..........................................................8-46 8.5 Inter-system Roaming Handover Optimization..............................................................8-52 8.5.1 Common Inter-system Handover Strategies.......................................................8-52 8.5.2 Current Huawei Inter-system Roaming Handover Strategies..............................8-53 8.5.3 Applied Inter-system Roaming Handover Strategies...........................................8-55 Appendix A Acron yms and Abbreviations...........................................................................A -1 Appendix B References...........................................................................................................B -1 WCDMA Radio Network Optimization Guide Chapter 1 IntroductionHuawei Technologies Proprietary1-1 Chapter 1 Introduction The Radio Network Optimization (RNO) is to reasonably adjust the communicationnetwork planning and design as per certain criteria so as to enable the networkoperation more reliable and economical, improve the network service quality andresource utilization, which is of great importance to network carriers.While questing for reasonable utilization of network resource and improving theservice quality to the greatest extent, the network optimization solution can provideevidences and principle for future network expansion. The network optimization cansupplement and correct the network planning that was well implemented but hassome aspects not considered, adjust and correct to network when the conditions(such as transmission environment and traffic) change. Therefore, in the wholesolution, the network optimization plays a key role.In general, drive test, statistics and users subjective sense are used as optimizationevidences, while signaling tracing and analysis play a crucial role in troubleshooting.The basic method for network optimization is to analyze an adjust engineeringparameters and cell parameter according to OMC statistics result and drive test resultso as to optimize the network finally. In this guide, the flow of network optimization will be introduced firstly, and then eachprocedure in the flow will be described in details together with actual cases. WCDMA Radio Network Optimization GuideChapter 2Radio Network Optimization Work FlowHuawei Technologies Proprietary2-1

Chapter 2 Radio Network Optimization Work Flow 2.1 Work Flow For the live network, there are OMC statistic data that can be used to reveal thenetwork performance.On the other side, the influence to the network should be taken into consideration forany adjustment. That is the reason that the adding of the simulated loading and thechange of the parameters must be done at night if necessary.The optimization process is divided into the following parts: Project preparation and startup In this stage, we need to prepare the resource needed by optimizing project, including engineers vehicles, software and tools, and we also need to divide the cluster if thenumber of sites is more than the specified scale. More details refer tochapter 3. Single site verification In this stage, firstly we check the alarm and states of target cells in RNC LMT. Thenwe need to verify the coverage, scrambling codes, installation of the target cells andmake one call test to verify the base function of target cells. More details refer tochapter 4. RF optimization In this stage, we need to optimize the coverage and handover performance and solvethe RF related call drop. More details refer tochapter 5. Service Optimization In this stage, we need to optimize performance of each service, including AMR, VP, and PS service. More details refer tochapter 6. Report output After the optimizing project completing, we need output the final optimization report. More details refer tochapter 7.

Figure 2-1 Radio network optimization work flow 2.2 Related Report 2.2.1 Output Report and Table Optimization Report Site Configuration Parameter Table Radio Parameter Configuration Table

2.2.2 Process Table Projectpreparationand startup Optimization databaseAnalysis andadjustmentVerificationtest Report output YN Analysis andadjustmentVerificationtest YN RF otimization Service otimization Single siteverification O t i m i z i n m e t h o d Site Configuration Parameter Table Radio Parameter Configuration Table

Adjust system parameter Aadjust engineering parameter Data Collection Drive test CQT Statistics Satisfy criteriaS a t i s f c r i t e r i a WCDMA Radio Network Optimization Guide Chapter 3 Project Preparation and StartupHuawei Technologies Proprietary3-1 Chapter 3 Project Preparation and Startup 3.1 Understanding of the Existing Network Before the network optimization, the RNO manager (RNO_PM) should be acquaintedwith the existing network to make the working plan.The information below is required: l radio network planning report l latest site configuration table and radio parameter configuration table l OMC statistic data l subscriber complaints of the existing network 3.2 Optimization Team Establishment and Cluster Division 3.2.1 RNO Team The optimization manager shall establish the optimization teams by estimating theworkload, deciding the human resource and tools requirements and making the timeschedule.The general organization is shown in Figure 3-1. WCDMA Radio Network Optimization Guide Chapter 3 Project Preparation and StartupHuawei Technologies Proprietary3-2 Figure 3-1 Organization of RNO team 3.2.2 Cluster Division If the NODEBs involved are more than 20 or the test time of the target area exceeds3 hours, we should divide the target area into parts before the optimization. Clusterrefers to a set of NODEBs in one part. An area of a cluster should contain a completetest route and the time for one test should be less than three hours. In addition, thetest time of each cluster should be similar. 3.3 Optimization Tools and Software 3.3.1 Optimization Tool Each team should be equipped with a set of drive test tools, including: l UMTS scanner: Agilent E6474A with E6455C(version upper 6.0) scanner or DTIscanner l

UMTS test UE: HUAWEI U626 or Qualcomm TM6200 (version upper 5.0) l GPS and data line: The E6455C which has built-in GPS.More than one set of equipment below is required and shared in a large project team:

WCDMA Radio Network Optimization Guide Chapter 3 Project Preparation and StartupHuawei Technologies Proprietary3-3 l YBT250: For identifying the location of the interference l Digital camera: Used to record special propagation environment 3.3.2 Optimization Software The following software is required for each team: l Post processing tool: HUAWEI Assistant or ACTIX Analyzer RVS l MapInfo: For some self-definition analysis of drive test data

WCDMA Radio Network Optimization Guide Chapter 4 Single Site VerificationHuawei Technologies Proprietary4-1 Chapter 4 Single Site Verification 4.1 Purpose The purpose of single site verification is by using Scanner + UE to: l verify coverage (RSCP & Ec/Io) l verify scrambling codes l verify no installation faults l Verify the base function 4.2 Preparation Before the time-consuming drive test, you should only ensure the following conditionshave been met. 4.2.1 Alarm Check Usually, the product support engineer will be responsible for checking and resolvingall the alarms, It is important for the optimization engineer prior to any field test tocheck with product support engineer regarding any alarms especially intermittentalarms. 4.2.2 Cell State Check Usually, the product support engineer will be responsible for checking and resolvingall the cell state problems. It is important for the optimization engineer prior to anyfield test to confirm with product support engineer regarding cell state. 4.2.3 Radio Parameters Check

Check equipment version and radio parameters configuration, for example: thescrambling code, power setting, neighbor cell list, handover parameter. As forneighbor list, the check must be done not only to intra-frequency neighbor list, butalso inter-RAT neighbor list.

WCDMA Radio Network Optimization Guide Chapter 4 Single Site VerificationHuawei Technologies Proprietary4-2 4.3 Site Verification Site verification is to check the service using one call test before the optimization toensure the basic services are normal for each site.The check consists of signal strength, feeder connection, LA/RA update, PS attach,PS detach, voice service call setup, PDP activation , intrafrequency handover, 3G/2Ghandover, 3G/2G cell reselection.If there is any suspect like the low signal strength and wrong feeder connection (forexample Tx/Rx main antenna or Rx diversity antenna), the optimization engineer shallask installation engineer to check the RF.

WCDMA Radio Network Optimization Guide Chapter 5 RF OptimizationHuawei Technologies Proprietary5-1 Chapter 5 RF Optimization Any RF issues at edge of cluster must be resolved in coordination with the adjacentcluster as soon as possible. 5.1 Purpose The purpose of RF optimization is by using Scanner + UE to: l optimize CPICH and service coverage and cell dominance l minimize interference l minimize pilot pollution l optimize neighbor list l resolve any RF-related drop call l improve connected ratio, delayed time connected 5.2 Drive Survey Drive survey includes both a scanner and a UE in a continuous AMR call. Drop callsshould be analyzed to determine if the cause is RF related or not. 5.2.1 Preparation I. Alarm Check Refer toAlarm Check II. Cell State Check

Refer toCell State Check III. Test Route Selection Usually, the customer can provide the test route based on GSM experience. IV. Load Simulator It s not recommended to do the test with simulated load for a live network becausethe simulated load will affect the users.

WCDMA Radio Network Optimization Guide Chapter 5 RF OptimizationHuawei Technologies Proprietary5-2 5.2.2 Drive Test Drive the test vehicle along the test route. Use the foreground data collecting tools torecord the corresponding data at the UE side. Use the RNC LMT to recode thecorresponding data at the RNC side. 5.3 Data Analysis The drive test data analysis consists of: l poor coverage l dominant cell l pilot pollution l neighbor list missing l minimize interference l other any RF-related drop call 5.4 Adjustment Recommendation and Implementation Changes will mainly include: l engineering parameter adjustment l neighbor list changes

I. Engineering Parameter Adjustment Most of the problems of coverage or interference can be solved by adjusting theengineering parameter. The purpose is to enhance the coverage to eliminate the blindarea or conversely to control coverage and interference. To be specifically, to use thecombination of ways below: l Antenna down tilt adjustment l Antenna azimuth adjustment l Antenna location adjustment l Antenna height adjustment l Antenna replacement l Site replacement l Add new cellBefore the implementation of the adjustment of the engineering parameters, it snecessary to carefully analyze the RSCP, dominant cell, Ec/Io of related cells.Usually the optimization engineer only needs to provide installation engineer with theengineering parameter change requirement.

II. Radio Parameters Adjustment As for radio parameter changes in RF optimization phase, usually only the neighborcell list is taken into consideration.All parameters should be adjusted very carefully. III. Benchmark Sometimes it s hard to predict accurately the impact of the parameter changes. So it snecessary to benchmark before and after parameter changes. It s recommended todo the benchmark together with the statistics analysis for live network.Another benchmark needs to be mentioned is for different type of UE.

IV. Summary After the RF optimization, the team leader (NO_PL) should output: l Site Configuration Parameter Table l Radio Parameter Configuration Table WCDMA Radio Network Optimization Guide Chapter 6 Service OptimizationHuawei Technologies Proprietary6-1 Chapter 6 Service Optimization Service optimization includes field test and statistics analysis.The field test can be divided into two parts: drive test on main road and test forspecial areas (CQT). Test for special areas is like walk survey for some indoordistributed cells. 6.1 Purpose The purpose of service optimization is by using scanner + UE log, RNC log, statisticscounters and KPI, MSC & SGSN log (if needed) to: l reduce access failure l reduce drop call l enhance service quality 6.2 Drive Survey Drive test is to get information of the service on the main road and to solove theproblems as access failure, drop call, power control, handover and service quality.The test includes (in order of priority): l The call setup test for voice service l The continuous call test for voice service l Idle mode test l The call setup test for video phone service l The continuous call test for voice service l

The call setup test for PS service l The continuous call test for PS serviceThe test should be performed repeatedly to the location with problems to make surethat the problem is repeatable.Unlike RF optimization drive test, for service optimization it s not needed to do theservice test for all sites. So it s not necessary to stick to the idea of cluster. 6.2.1 Preparation Refer toPreparation WCDMA Radio Network Optimization Guide Chapter 6 Service OptimizationHuawei Technologies Proprietary6-2 6.2.2 Drive Test Refer toDrive test 6.3 Data Analysis The data analysis consists of: l Analyzing and finding out the solution for access failure (system bugs, powercontrol parameter, cell selection and reselection parameter etc.) l Analyzing and finding out the solution for drop call (system bugs, power controlparameter, handover parameter etc.) l Analyzing and finding out the solution for service quality related problems(system bugs, power control parameter, RLC parameter etc.) 6.4 Adjustment Recommendation and Implementation The adjustment recommendations for service optimization are mainly focus on theradio parameters. I. Radio Parameter Adjustment The parameters to be changed usually are: l common control channel power allocation l RL maximum power l intra-frequency handover

l inter-frequency handover l Inter-RAT Handover l power control l Access parameters l Other related parameters.The influence caused by the cell parameters adjustment should be analyzed carefullyby optimization engineer.The parameters adjustment to a live network should be performed when traffic is low(for example: at midnight). It is essential to do the service functional test afteradjustment. II. Benchmark Refer toBenchmark WCDMA Radio Network Optimization Guide Chapter 6 Service OptimizationHuawei Technologies Proprietary6-3 6.5 Test for Special Areas (CQT) Generally, special area refers to a small district or indoors environment, such as: l operator offices, residence of VIP l key hotels or entertainment location l government location l locations of large company or group subscribers l locations tend to be cared , such as railway station or airport 6.6 Statistics Analysis As for the live network, in parallel with field test, you need to optimize the cells whoseperformance indicators have not reached the acceptance requirements. For example,to optimize the cells with congestion problems in busy hours. 6.6.1 Preparation Check with engineers from operator OMC department and operator QA department: l

all necessary counters are activated l the KPIs have been created l tools and scripts for KPI analysis are available. 6.6.2 Analysis Statistics analysis includes:1) identify the poorly performing RNCs, sites and cells2) finding out the cause of the degrading together with other engineers (productsupport engineer, transmission engineer, installation engineer)3) give information about the impact of the changes of the parameters forbenchmark. 6.6.3 Adjustment Recommendation and Implementation Usually it refers to the radio parameter adjustmentRefer toAdjustment Recommendation and Implementation 6.7 Summary After the parameter optimization, the team leader (NO_PL) should output: l Radio Parameter Configuration Table WCDMA Radio Network Optimization Guide Chapter 7 Output ReportHuawei Technologies Proprietary7-1 Chapter 7 Output Report 7.1 Optimization Output Report and Table The optimization report should be presented after the network KPI fulfilling theoptimization target.The optimization report should include: l project background introduction l existing network status l optimization target l optimization process l problem analysis and adjustment recommendation, and the effect for theimplemented part

l acceptance test and result l leftover problems and suggestions from the optimization view. l appendixBesides the optimization report, the updated Site Configuration Parameter table and Radio Parameters Table are required. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-1 Chapter 8 Category of Optimization Problems andCase Analysis 8.1 Coverage Optimization 8.1.1 Classifying and Defining Coverage Problem I. Signal Dead Zone A signal dead zone generally refers to the area where the pilot signal is less than thelowest access threshold (such as RSCP threshold is 115dBM, Ec/lo threshold is-18dB), and the area can be a concave area, hill rear, elevator shaft, tunnel,underground garage or basement, and internal part in a tall building.In general, if the coverage of adjacent base stations is not overlapping and there areconsiderable users or the area is large, a new base station should be established orthe coverage should be expanded (such as, improving the pilot transmitting powerand the antenna height at the cost of the capacity) to enable the overlapping depth toreach about 0.27R (where R refers to the cell radius), ensuring a certain-sized softhandover area.Meanwhile, the adjacent frequency interference possibly caused by expandedcoverage should be noted. In the concave area and hill rear, a new base station orRRU or repeater can be set up to efficiently cover the dead areas, or expand thecoverage. However, because the RF repeater may cause the inter-modulationinterference, the possibly generated interference should be noted during theengineering. For the signal dead area in the elevator shaft, tunnel, undergroundgarage or basement, and tall building, it can be solved with RRU, repeater, indoordistribution system, leaky cable and directional antenna. II. Coverage Hole A coverage hole refers to the coverage area where the pilot signal is lower than thelowest value required by full-coverage services such as voice, VP and PS 64Kservice, but higher than the lowest access threshold of mobile phones.For example, if the traffic is even but the sites locations are unevenly distributed,RSCP will not be able to meet the lowest requirement for full-coverage services insome areas. Still, if RSCP of the pilot signal in some areas can meet all requirements, WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-2 but the intra-frequency interference increases, the pilot channel Ec/lo will not meetthe lowest requirement for fullcoverage services.For example, because the cell breath effect happens for the increasing capacity of thecells around the soft handover area, the coverage quality in the soft handover area islowered, that is, a so-called coverage hole appears in the area. Here, the coveragehole is for mobile phone services, and is different from the signal dead area in whicha mobile phone usually cannot camp on a cell nor initiate location update and locationregistration and therefore causes Disconnected

.In general, the unreasonable site distribution should be avoided in the planning.Selecting a suitable location can guarantee either that the pilot RSCP density of thenetwork reaches a certain level, such as -65dBm in the road in a dense urban areaand -80dBm in an ordinary area, or that the pilot Ec/lo of the network with a certainloading is not lower than the lowest requirement for full-coverage services.Because of the restriction of estates and equipment installation, improper locationshave to exist. When a coverage hole happens, a new mini-base station or a repeatercan be set up to improve the coverage. If the coverage hole is not serious, thecoverage can be optimized by using high gain antennas, increasing antenna heightand reducing the mechanical tilt angle of the antenna. If the pilot Ec/lo coveragecannot be efficiently improved through RF adjustment, a dominant cell can begenerated by adjusting the pilot frequency (increasing the strongest and reducing therest). III. Trans-Regional Coverage The trans-regional coverage usually refers to the coverage of some base stationsexceeds the planned scope. The trans-regional coverage can cause discontinuousdominant areas in the coverage of other base stations if the strength of trans-regionalsignal meets the requirement for the full-coverage services.For example, some sites of which the locations are much higher than the averageheight of around buildings can transmit signals very far along highlands or roads, andform the dominant coverage, the so-called Island , in the coverage of other basestations. Therefore, when a call is accessed to the Island area which is far from anearby base station and the nearby cells are not set as adjacent cells to Island theduring setting the cell handover parameters, the call drop will happen as soon as themobile phone leaves the "Island .Even if the adjacent cells are configured, but for the too small Island area, the calldrop will happen if the handover cannot proceed in time.Still, in some areas like that on both sides of Victoria Harbor in Hong Kong, if aspecial planning is made for the base stations in Tsimshatsui in Kowloon, and the WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-3 Center and Sheung Wan in Hong Kong island, the trans-regional coverage betweenthese two areas will be caused due to the too near harbor sides and the interferenceis generated.In general, for the trans-regional coverage, the signal transmission of facing roadshould be avoided or using the obstacle effect with around building to reduce thetrans-regional coverage. But it should also be noted whether the intrafrequencyinterference is generated to other base stations. For the base station in high location,the effective method is to change the

location. However, because of the restriction ofestates and equipment installation, if a suitable location cannot be found, theintensive adjustment to the mechanical tilt angle of the antenna will cause that theantenna directional diagram is distorted. If necessary, the pilot power can be adjustedor the electric tilt angle antenna can be used so as to remove the "Island effect byreducing the coverage. IV. Pilot Pollution The pilot pollution means that too many pilots are received on a point, in which therehas not a dominant pilot.In this guide, the following method is used to judge the existent of the pilot pollution:the number of the pilots that meet the condition dBm RSCPCPICH 95_ > ismore than 3, with dB RSCPCPICH RSCPCPICH thst 5)__( 41 T302 N302.After T314 times out, theRB of the service corresponding to this timerwill be deleted.The suggested value: D20.T315 Timer315The value range is: D0, D10, D30, D60,D180, D600, D1200, D1800The physical range: 0, 10, 30, 60, 180, 600,1200 and 1800Physical unit: sContents: Timer315 is started when the radiolink failure rule is observed and only the radiobearing related to Timer315 exists. The timeris stopped when the cell updating iscompleted. The default value is 180.T315 (or T314) is started and the CELLUPDATE signaling is sent when the radio linkfailure happens to the users in theCELL_DCH. Before T315 (or T314)corresponding to the service times out, if there-configuration of the radio link configured byCELL UPDATE CONFIRM fails, the CELLUPDATE signaling can be re-transmitted toreconfigure the radio link, (related to T302and N302). In this way, the radio link can beconfigured again. For this purpose,T314>T302 N302.After T315 times out, theRB of the service corresponding to this timerwill be deleted.Suggested value: D30.N315 Constant315The value range is: D1, D2, D4, D10, D20, WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-37 Parameter ID Parameter name Parameter Description D50, D100, D200, D400, D600, D800, D1000The physical range is: 1, 2, 4, 10, 20, 50, 100,200, 400, 600, 800, 1000Physical unit: sContents: it represents the maximum times ofthe continuous synchronization indicatorreceived from L1 during Timer313 startup.The default value is 1.Suggested value: D1. 8.3 Access Optimization 8.3.1 Classified Definitions of Access Failure The data analyzed tool Assistant defines access failure as the rule that any situationwhere one of the following conditions is met after the original UE issues the RRCConnection Request:1) UE receives the RRC Connection Reject message;2) After having received the RRC Connection setup message, UE receivesor sends the RRC Connection Release message;3) During the Call setup process, UE receives the BCCH message;4) Timer times out, that is, within 3 seconds (T300) after UE has sent theRRC Connection Request message, UE fails to receive the RRCConnection Setup message.The data analyzing tool TEMS defines access failure as per the following rule (forvoice service): 1) Random access failure: After dialing, the RRC Connection Requestmessage is not sent;2) The RRC Connection Setup message is not received: After having sentthe RRC Connection Request message, UE fails to receive the RRCConnection Setup message.3) The RRC Connection Complete message fails to be sent: After having theRRC Connection Setup message, UE fails to send the RRC

ConnectionComplete message.4) UE receives the RRC Connection Reject message: UE receives the RRCConnection Reject message and fails to re-send the RRC ConnectionRequest message as a try. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-38 5) UE fails to receive the measurement control message: After having sentthe RRC Connection Complete message, UE fails to receive themeasurement control message.6) Failure to send CM Service Request: After having received themeasurement control message, UE fails to send the CM Service Requestmessage.7) UE receives the Service Request Reject message: UE receives theService Request Reject message.8) UE fails to receive the Call Proceeding message: After having sent theCC SETUP message, UE fails to receive the Call Proceeding message.9) UE fails to receive the RB Setup message: After having received the CallProceeding message, UE fails to receive the RB Setup message.10) UE fails to send the RB Setup Complete message: After having receivedthe RB Setup message, UE fails to send the RB Setup Completemessage.11) UE fails to receive the Alert or Connect message: After having sent theRB Setup Complete message, UE fails to receive the Alert or Connectmessage.12) UE fails to send the Connect Acknowledge message: After havingreceived the Alert or Connect message, UE fails to send the ConnectAcknowledge message.To sum up, any of the signaling failures before conversation will cause access failure. 8.3.2 RRC Connection Setup Failure Analysis The RRC connection setup failure can be detected by the UE signaling process andRNC single-user trace. RRC connection is established via the following steps: UEsends the RRC Connection Request message via RACH, RNC sends the RRCConnection Setup message via FACH, then UE establishes the downlink dedicatedchannel, synchronizes it, and now sends the RRC Connection Setup CMP messagevia the uplink dedicated channel. Reasons for RRC connection failures include: Errors or faults in uplink RACH,downlink FACH power distribution, cell parameter reselection, low downlink initialtransmitting power, uplink initial power control, congestion, abnormal equipment, etc,especially those of uplink RACH, downlink FACH power distribution, cell parameterreselection, and abnormal equipment. l The RRC Connection Request message sent by UE is not received by RNC. IfEc/Io of the downlink CPICH is not too low (say, about 12dB), then normallyRACH fails due to incorrect power control estimation for the uplink open loop, orinsufficient power escalation of Preamble, or lower-than-required output power ofUE. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-39 l RNC, after having received the RRC connection request sent by UE, sends theRRC connection Setup message but UE fails to receive it. Now check the CPICHEc/Io. If it is blow -12dB and there is no better cell in the monitored set, then itmust be failure of coverage, in which case, increase the FACH power. If there isa better cell in the monitored set, it might be caused by cell reselection, in whichcase, adjust the cell reselection parameters to speed up the reselection. l After having received the RRC Connection Setup message, UE fails to send theSetup Complete message. If the downlink signal quality is normal, then it mightbe caused by the abnormal mobile phone, or the initial power of the downlinkdedicated channel is too low to enable synchronization. l UE has sent the RRC Setup Complete message but RNC fails to receive it.Since the uplink initial power control will make the UE transmitting powerescalate, such failure is rare. If it does happen, increase as appropriate theConstant Value of the dedicated channel. 8.3.3 Paging Failure Analysis Paging failure occurs if the terminal call receives the direct transfer message ofDisconnect sent from CN, as shown in Figure 8-23.Judged from the UE signalingprocess, no abnormality can be found in the called party, although there have beentimes that UE fails to send the RRC connection setup request after having receivedthe Page message. From the RNC single-user tracing of the called party, it can befound that the terminal call has received the Page message sent by CN but no furthermessage from then on.

Figure 8-23 Signaling process of the calling UE WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-40 Causes for paging failure mainly include: Failure of RNC to issue the Page message,lower-than-required power of the paging channel or the paging indication channel,occurrence of cell reselection on UE, etc. 8.3.4 Access-related Parameter Configuration I. FACH Transmitting Power This parameter sets the FACH transmitting power. If it is too low, the UE at the celledge will not be able to correctly receive the service and signaling carried by FACHand thus affect downlink common channel coverage, which will eventually affect thecell coverage. If it is too high, FACH will interfere with other channels and occupy thedownlink transmitting power, which will affect the cell capacity. In Huawei baseline,the FACH power offset is 1dB, is set on the basis of the cell edge CPICH with Ec/Io at-12dB. If the field coverage is even poorer, the FACH power should be increased inlight of the Ec/Io value of the cell edge CPICH. II. PCH Transmitting Power This parameter sets the PCH transmitting power. If it is too low, the cell edge UE willnot be able to correctly receive the paging message and the paging delay will beincreased, and thus bring down the paging success rate, which will eventually affectthe access success rate. If it is too high, power will be wasted and downlinkinterference will also be increased. III. PICH Transmitting Power This parameter sets the PICH transmitting power. If it is too low, the cell edge UE willnot be able to correctly receive the paging indication message, the paging delay willbe increased and error might occur in reading PCH, which will waste UE batterypower, affect the downlink common channel coverage and eventually affect the cellcoverage. Since PICH transmission is continuous, if the transmitting power is too high,PICH will interfere with other channels, occupy the downlink transmitting power andaffect the cell capacity. Therefore, it is recommended not to increase the PICHtransmitting power. To increase the PICH coverage, you may reduce NP to 18, sincethat value is enough for the initial stage of the network and it is actually the typicalconfiguration in practice, although NP reduction will reduce the paging capacity of theUu interface. IV. Cell Reselection Parameter Qhyst2s According to Rule R, the measurement of the current service cell, added withQhyst2s, participate in the cell reselection sequencing. Value of the parameter WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-41 depends on the slow fading of the area where the cell is located. This parameterfunctions to prevent the pingpong handover, caused by slow fading, of the cellreselection results when the UE is at the cell edge. The pingpong handover maycause frequent location updating (in the idle mode), URA updating (URA_PCH) or cellupdating (CELL_FACH, CELL_PCH) which will add to the network signaling load andUE battery consumption. V. Cell Reselection Parameter Treselections

If the signal quality (the CPICH Ec/No value measured by UE) of other cells isconstantly better than that of the current cell within the period designated by thisparameter, UE will reselect the current cell as the resident cell. The parameter usesthis function to prevent UE to perform the pingpong handover between cells. VI. Cell Reselection Parameter Sintrasearch This parameter is the threshold for initiating the intra-frequency measurement. Whenthe local cell Ec/Io value is lower than QRelxmin+2*Sintrasearch, the intra-frequencymeasurement will be initiated. This parameter will affect the cell reselection speedand eventually the one-whack UE access rate and IU port paging success rate. Withrelatively smaller influence ensured on the UE power consumption, it isrecommended to set the parameter as high as possible. VII. Cell Reselection Parameter Qoffset2 Values of the signal quality of the adjacent cells have to have Qoffset2 deductedbefore participating in the Rule R appraisal. For ordinary single-layer cell, theparameter can be set to 0 and the same effect can be achieved via Qhyst. Therefore,normally no adjustment is recommended. VIII. AICH Transmitting Power If set too low, the parameter will cause the cell edge UE to fail to correctly receive thecapture instruction and thus affect the downlink common channel coverage. InHuawei baseline, the parameter is set to -6dB. Judged from the optimization result,the AICH transmitting power is sufficient for downlink coverage, and as AICHtransmission is continuous, it will occupy relatively large downlink capacity ifincreased. IX. PRACH Related Parameter In light of the uplink PRACH failure, related parameters of the channel should beadjusted, which include the number of preamble retransmission attempts, incrementsof preamble power escalation, preamble, Message and power variance, etc. The WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-42 above parameters are interdependent. In case of the PRACH failures, it isrecommended to adjust the preamble re-transmission attempts from the currentbaseline value 8 to 20 to avoid PRACH failures. Typical case analysis Repeated RRC Connection Request caused by cell reselection 1) Description of the case Figure 8-24 Signaling at the UE side As shown in Figure 8-24, the interval between the two RRC Connection Requests byUE is about 1.2S. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-43 Figure 8-25

Signal quality at the first RRC Connection Request by UE WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-44 Figure 8-26 Signal quality at the second RRC Connection Request by UE 2) Analysis of the case As the baseline, Treselection is set at 1, Qhyst2 at 2dB, Qoffset2 at 0 dB andSintrasearch at 5. Even if the signal of the target cell is better than that of thelocal cell, cell reselection cannot complete until 1 second later. Therefore, if thesignal variance between the target cell and the local cell is same as the abovedescribing, there will be little margin for optimizing the cell reselectionparameters. If Treselection is set at 0, then the time needed for reselection willbe 8*DRX, which far exceeds 1 second since the lowest possible DRX is 0.64second. Still, if Treselection is set at 0, then the protocol requires Ec/Io of thetarget cell to be 3dB than that of the local cell, and the period for completing cellreselection calculated on the basis of multiple measurements will be 1.2~1.4seconds. 3) Solution to the caseTo shorten the period for cell reselection, we once set Qhyst2 to 0 andSintraSearch to 7 and found that, in walk tests, pingpong handover may occurand the period for cell reselection is not shortened. Therefore, it is recommendedto maintain Qhyst2 at 2dB and set SintraSearch in such a manner as to causeUE to initiate the intra-frequency measurement as early as possible. With WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-45 relatively small UE power assumption ensured, it is recommended to setSintrasearch to 7. 8.4 WCDMA Network Uplink Interference and Optimization In the WCDMA network building, we find that there exists widespread uplinkinterference, slight or severe, in the WCDMA network. On the one hand, since theWCDMA network is basically built after the 2G network is well developed.Acquisition of the sites, equipment room space and roof space is difficult, which,together with consideration for saving the cost of network building, causes manycarriers to build networks with the antenna feeder shared by 2G and 3G. In currentpractice of the WCDMA network building, because the project quality cannot be madegood enough, the inter-modulation caused by antenna feeder connector, adapter andbearer when the antenna feeder is shared by 2G and 3G is constant. Suchinter-modulation is generally categorized by us into internal interference; On the otherhand, mobile communication is evolving from the second to third generation, newtechnologies mushrooms, and new mobile communication carriers are rapidlydeveloping, which together causes the radio frequency resources to be more andmore scarce and various potential sources of interference produced at an amazinglyspeed. The former dedicated radio system occupies the existing frequency resources,network configuration by different carriers is improper, transmitters are incorrectlyconfigured, together with cell overlapping, environment and electromagneticcompatibility and intentional interference, these are all causes for radio frequencyinterference for the mobile communication network. All currently available mobilecommunication systems occupy radio frequencies of 2.5G and below. For such aband, mutual interference is inevitable. Therefore, there do exists the radio frequencyinterference in the mobile communication network. Such interference is categorizedby us into the external interference. Internal or external, the interference will increasethe noise level of the base station. Since WCDMA is a self-interference system, uplinkinterference will necessarily cause reduction in the uplink coverage. With severeuplink interference, the effective mobile phone signals are immersed and thus cannotbe demodulated, which results in access failure or call drop. As the uplink interferenceaffects the users of whole base stations, it is much more severe than downlinkinterference which affects just one or several users, and this is also the reason whywe pay so much attention to uplink interference of the WCDMA system. As per 3GPP,all NodeBs are capable of testing RTWP, such RTWP measurement function of theNodeB is an extremely important device for us to detect interference. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-46 8.4.1 Cause Analysis of the Uplink Main Diversity RTWP Variance Figure 8-27 Variance of the uplink main diversity RTWP As shown in Figure 8-27, the main diversity RTWP has obvious variance. By analysis,the causes may be:

l Intensity of the interferences on the main diversity has much variance; l Some cells such as the indoor base stations use only one receiving channel; l NodeB has never performed the radio frequency channel calibrating. 8.4.2 Cause Analysis of the RTWP Abnormality I. Uplink Inter-modulation Interference 1) Causes for inter-modulation interference Inter-modulation generally is caused by multiple frequency signals. But when there isonly one non-constant amplitude, new frequency components might also be produced.That is, in the non-linear system, a single amplitude modulation signal may alsoproduce new frequency components, which is the cause for spectrum expansion andwe deem it the inter-modulation product. In case of poor connection, even the CWsignal may produce new frequency components, which is also deemed by us to beinter-modulation. Here we have expanded inter-modulation: If there emerge newfrequency components in the system, inter-modulation occurs. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-47 Inter-modulation is generally caused by different characteristics of different voltagesof active devices. But actually even passive devices with a large power may alsoproduce inter-modulation, which is called by us the passive inter-modulation. Passiveinter-modulation: Intermodulation caused when two or more signals are mixed in thenon-linear passive devices (such as the duplexer, isolator, antenna or cable joint).Most of the inter-modulation we encounter is passive inter-modulation. How does ithappen? Firstly, it can be caused by unstable connection between the base stationand the antenna side. For example, too loose or too tight connector DIN and thescrew problem can both causes inter-modulation; secondly, there exist impurity in the joints, as must be caused by improper construction; and thirdly, there exists impurityin the point of weld in the antenna connector, or impurity in the antenna feeder point,or there exists a rosin joint. 2) Characteristics of inter-modulation interference l The fluctuation of RTWP is generally high, at about 10dB or even higher; l There exists a certain relationship between interference and traffic. With lowtraffic, there is a high probability of absence of interference; l Interference happens in a cluster and sustain for a period of time without abruptchanges, which is dramatically different from external interference; l RTWP fluctuation is not in cycles; l RTWP of the main diversity are not related. If they become related, there mustbe special reasons. For example, they are somehow combined;RTWP fluctuation with several of the above characteristics can be caused byinter-modulation. If it has the above all five characteristics, it can be nothing butinter-modulation. 3) Solution to inter-modulation interference

l Check the roof if there exists any metal article around the antenna or theantenna is too close to the antennas of other carriers; l Check the DCS bearer characteristics to find the 2G frequencies that combinewith the 3G frequencies, and mark out the BCCH branch; l Disconnect the respective jumpers on the NDDL branch of the main diversity ordiversity to be tested at the NodeB side, and connect them to YBT. Now checkthe frequencies received by YBT and record the frequencies and theiramplitudes; l Utilize the frequencies received by YBT to perform analysis on inter-modulationpossibility, then connect the jumpers back to NDDL. If BCCH is not on thebranch in question, re-connect it to the branch in question; l Slightly strike on each of the connectors on the branch in question to observeRTWP. If it has high fluctuation, then the connector must be faulty and should betreated by engineers; l If, after the above test, still nothing wrong is found in the connector but theinterference persists, then use YBT+ filter+ antenna to test the frequency WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-48 characteristics of the interference received by the WCDMA antenna. Now useYBT+ filter+ directional antenna to test the interference frequency spectrum inthe air. If the directional antenna detects interference in the air, check the sourceof interference. If the interference comes from the metal structure in front of theantenna, then it is radiation-caused passive inter-modulation, which can besolved by the customer adjusting the antenna or the environment around it; l If inter-modulation interference is detected, normally the 2G frequencyconfiguration shall be checked, as judged from the experience of locating thebase station with inter-modulation interference. If the configuration of the 2Gfrequency that combines with the 3G frequency directly causes the latter toplummet into the WCDMA receiving band (for example, the 5xx and 8xxfrequencies co-exist in the 2G configuration), then such configuration should berevised as much as possible. Typical case analysis Multiple-frequency inter-modulation caused by load Figure 8-28 RTWP 24-hour measurement inter-modulation caused by load The cell in this case is an indoor system, with combined output of 2G signals of twocarriers. RTWP is abnormal in the cell. The field study finds that it is caused by looseload connection. A slight touch on the load will trigger violent RTWP fluctuation, whichdisappears after the load is replaced.

WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-49 II. Uplink External Interference 1) Classification of frequently seen external interference Frequent external interference includes: repeater self-excitation interference onthe ambient sites, donor cell uplink interference caused by repeaterself-excitation, donor cell uplink interference caused by improperly configuredrepeater gain, uplink interference caused by the close-distance radiation of the2G repeater antenna on the 3G antenna, intermittent interference on the sitesnear the railway caused by passing trains, uplink interference caused by indoorair-conditioner remote control switch, uplink interference caused by the switchon/off of the indoor emergent light, etc. 2) Characteristics of external interference l Judged from the time-related characteristics of the RTWP data, externalinterference does not last long. Normally it disappears within several toseveral dozen seconds and will not reappear until a long period later; l Normally, the interference is related to the main diversity; l Normally, the interference affects multiple cells; l Normally, the interference occurs when the interference source has arunning status change.3) Solution to the external interference l Collect basic data and perform detailed analysis prior to going to the siteSuch analysis includes: analyzing the long-term characteristics of the RTWP datacollected in various periods, analyzing the short-term characteristics of the RTWPdata collected in various periods, analyzing the long- and short-term characteristics ofthe RTWP data of the target cell to be located and its adjacent cells in a certain period,analyzing the environment of the target cell, finding the distribution information of therepeaters around the target cell, analyzing the relatedness of the main diversity of thetarget cell in light of the antenna configuration, predicting the interference sourcelocation on the basis of analyzing the RTWP data of the target cell and adjacent cells,the antenna direction and height. l Go to the site for interference source locationGo to the roof and check the environment; Find the relative locations of the base stations and the actual environments; Test the frequency spectrum and direction of the interference; Lock the suspect equipment on the basis of the analysis result and the interferencefrequency spectrum characteristics; Select two test points of the suspect equipment and test their specific strength of theinterference signal. Based on the change of the signal intensity, find the roughlocation of the interference source; WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-50 Go to the suspect location of interference, ascertain the location and the specificinterference source (to ascertain the interference source, you may utilize therelationship between interference and the status changes of the equipment such asswitch on/off, power on/off, etc). Typical case analysis

Uplink interference on the adjacent sites caused by the repeater self-excitation A trial found that, during several days, some sites had abnormal RTWP datasimultaneously, as shown below: Figure 8-29 Abnormal RTWP data in site 2 interference caused by the repeaterself-excitation WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-51 Figure 8-30 Abnormal RTWP data in site 3 interference caused by the repeaterself-excitation Figure 8-31 Figure 34 Abnormal RTWP data in site 4 interference caused by therepeater self-excitation The interference occurs regularly. That is, it occurs once an hour and affects severalsites. Check of the site distribution in the area around the interfered site finds that WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-52 there is a repeater in the center of the sites. By analysis and judgment, the repeatermust produce a self-excitation signal once an hour, which will affect the uplink signalof the adjacent cells. After the repeater problem is solved, all RTWP data of the sites become normal. 8.5 Inter-system Roaming Handover Optimization 8.5.1 Common Inter-system Handover Strategies I. Handover Based on Coverage To ensure continuous coverage of the GSM/WCDMA users in areas where there is no3G system coverage, the coverage-based handover or cell reselection is required forcontinuity of service. In special cases, there may be only 3G coverage in some areas. GSM also supportsthe same function to ensure the inter-system handover from GSM to 3G and cellreselection, and ensure continuity of service without GSM coverage. II. Handover Based on Load Load-based handover or cell reselection is applied for load sharing between GSMand WCDMA systems. The trigger for such a handover is the measured cell load ofGSM or the WCDMA network. For example, no matter how much load the GSM network has, hand over thenon-apparent CS data to the WCDMA network to ensure better throughput; when theGSM network has a load too heavy, now hand over part of the voice and apparentdata to the WCDMA

system. We should apply the load-based handover in the GSMnetwork to relieve the high pressure of the GSM network while saving the carrier sextra investment on the GSM network.In the packet domain, to realize the cell reselection by reason of service and load, thecell reselection should be controlled by the network. That is, the network cancommand the UE to change the cell or the system.The intersystem handover further enhances the load-based handover and cellreselection. The purpose is to utilize the two systems to the best. That is, when one ofthem has free capacity, the handover can avoid excessive investment on the other.Load equilibrium can also enlarge the system capacity. The total system capacity willbe larger than the sum of the two. WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-53 III. Handover Based on Service Service-based handover is applied to guide the traffic of the GSM or WCDMA network.Even when there is no load in the source cell, the service-based handover can alsoproceed. During the load-based handover/cell reselection or call setup period, theservice-based handover can be utilized. With this function, carriers can guide thevoice service to GSM and all VP and packet data to WCDMA to reduce investment onthe 3G circuit network and 2G packet switched network.The service priority information is stored in the core network (MSC/VLR or SGSN).During the call setup, utilize Iu_PS, Iu_CS or A and Gb interfaces to notify thewireless access part of the GSM and 3G network.Complete service handover is performed jointly by the access network and the corenetwork, where the core network commands, via the service handover configurationof available cells in the RAB assignment message, the access network to hand overthe service to the GSM network, or commands it to remain in the 3G network orissues no command at all. As the protocol, there can be the following three options:Handover to GSM should be performed. In this case, the service should be handedover to the GSM network;Handover to GSM should not be performed. In this case, the service should not behanded over the GSM network. But because of coverage and other reasons, theservice can still be handed over to the GSM network; and Handover to GSM shall not be performed. In this case, the service simply cannot behanded over to the GSM network. If the core network performs the above-described inter-system service handover, thenthe core network will have more flexibility in controlling the inter-system handover. Forthe core network, it is optimal to hand over all services as appropriate to the GSMnetwork so that the carrier can fully utilize the 2G equipment. 8.5.2 Current Huawei Inter-system Roaming Handover Strategies To reduce influence on the existing GSM network, improve the handover success rate,avoid the pingpong handover and cell reselection, and lessen signaling exchange, thefollowing inter-system handover strategies are recommended: l For the voice service, handover from WCDMA to GSM is supported while thereverse handover is not: For GSM carriers, the GSM coverage is already very good and it is thus unnecessaryto handover a conversation in process in the GSM network to the WCDMA network.Therefore, the GSM network does not need upgrading, and the pingpong handover is WCDMA Radio Network Optimization GuideChapter 8Category of Optimization Problems and Case AnalysisHuawei Technologies Proprietary8-54 thus avoided and signaling exchange between the WCDMA and GSM networksreduced. When the dual-mode UE resides in the WCDMA network and in conversation, movingfrom the GSM/WCDMA converging area to purely GSM coverage area, and nowreaches the border of the WCDMA network, the WCDMA network will initiate thehandover from WCDMA

to GSM according to the measurement report. The GSMnetwork outside the border should support the inter-system handover from WCDMAto GSM. When the dual-mode UE resides in the GSM network and in conversation, movingfrom the purely GSM coverage area to the GSM/WCDMA converging area, thenetwork should take measures to avoid handover from GSM to WCDMA. When theconversation is finished and UE is in idle, it can now reside in the WCDMA networkvia PLMN reselection or cell reselection. If only the unidirectional handover from WCDMA to GSM and the switch back toWCDMA via PLMN reselection or cell reselection are to be supported, then the GSMnetwork does not need upgrading. l For data service, handover is supported from WCDMA to GSM/GPRS via cellreselection: When the dual-mode UE resides in the WCDMA network and in conversation, movingfrom the GPRS/WCDMA converging area to purely GPRS coverage area, and nowreaches the border of the WCDMA network,