Neighbour Optimization RU10_Nokia

3G RANOP2 RU10 Module 3-Neighbour & RF Optimisation

For internal use 1 ©N No okia Siem ens Networks

Presentation / Author / Date

Module Mod ule 3- Nei Neighb ghbour our & RF Opt Optimiza imizatio tion n Objectives After this module the participant shall be able to: •Describe the main neighbour optimization methods and tools •Describe KPI analysis for neighbour list optimization •Learn how neighbou neighbourr & RF optimization optimization can be done with with Optimizer2. Optimizer2.0 0



Module Mod ule 3- Nei Neighb ghbour our & RF Opt Optimiza imizatio tion n Objectives After this module the participant shall be able to: •Describe the main neighbour optimization methods and tools •Describe KPI analysis for neighbour list optimization •Learn how neighbou neighbourr & RF optimization optimization can be done with with Optimizer2. Optimizer2.0 0



Contents • Neighbour optimization Methods & Tools • Adjacency Based Measurements • NetAct Optimiser Optimiser 2.0 overview overview • Propagation Delay



Neighbour optimization •

Handover Handover measurement measurement provides provides cell cell to cell HO statistic statistic which can be used for neighbour performance optimisation 

1013 Autodef SHO



1014 Autodef IFHO



1015 Autodef ISHO

•

Measurement Autodef HO Measurement

can can be used and activated activated as any any other basic basic RAN performance measurements

•

Autodef HO measurement

are also used with help of the NetAct optimiser 2.0 optional feature (automatic adjacency optimisation). •

NetAct Optimiser 2.0 use Autodef counters for deleting adjacencies and ICSU statisti statistics cs for creating creating adjacencie adjacencies. s.



Neighbour list Combination procedure- SHO/ISHO to undefined neighbour possible I •

Active Set may contain cells, which are not necessary adjacencies with each other.

•

The list of cells to be measured is send by the RNC in a MEASUREMENT CONTROL message and is changed at every Active Set Update. The RNC then combines the Neighbour lists according to the following rules: 1. Active set cells are included 2. Neighbour cells which are common to three active set cells are included 3. Neighbours which are common to the controlling cell and a second active set cell are included. (cell, other than the controlling cell, which has the highest CPICH Ec/Io) 4. Neighbour cells which are common to two active set cells are included 5. Neighbour cells which are defined for only one active set cell are included 6. Neighbours which are defined only for the second ranked cell are included 7. Neighbours which are defined only for the third ranked cell are included

•

If the total number of cells to be measured exceeds the maximum value of 32 during any step then handover control stops the Neighbour list generation

For internal use 5 © Nokia Siemens Networks


Neighbour list Combination procedure

II

• Because of the combination explained in the previous slide, it is possible to measure handover activity between 2 cells which do not have an adjacency defined between them. • In this example intra-frequency adjacencies exist between cells 2-6 and 6-7, but not between 2-7. Activity is measured when the lists of cells 2 and 6 are combined and 7 can be added, while 2 is still the best cell in the Active Set. The same effect applies for InterSystem list combining Neighboured

1 3

2

4 5 6

Not neighboured 9



7 8

UE path

Active Set Add FAIL_UE/NO REPPLY- example (wcel level) 25000

30

25 20000

20 15000 AS_UPDATE_RL_ADD_ATT 15

AS_UPDATE_RL_DEL_ATT AS_UPDATE_RL_ADD_FAIL_UE AS_UPDATE_RL_ADD_NOREPLY

10000 10

5000 5

0

0 1

18 35 52 69 86 103 120 137 154 171 188 205 222 239 256 273 290 307 324 341 358 375 392 409



Active Set Add/Del success- example, RNC 0.9999

0.04%

0.9998

0.04%

0.9997

Some failures due to no reply to ASU

0.03% 0.9996 0.03%

0.9995

AS_UPDATE_RL_ADD_SUCC % 0.9994

0.02%

AS_UPDATE_RL_DEL_SUCC % AS_UPDATE_RL_ADD_FAIL_UE % AS_UPDATE_RL_ADD_NOREPLY %

0.9993

0.02%

0.9992 0.01% 0.9991 0.01%

0.999

0.9989

0.00%

8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 0 0 2 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . . 3 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 7 . 1 8 . 1 9 . 2 0 . 2 1 . 2 2 . 2 3 . 2 4 . 2 5 . 5 . 6 . 7 . 8 . 9 . 1 0 . 1 1 . 1 2 . 1 3 . 1 4 . 1 5 . 1 6 . 1



Content • Neighbour optimization Methods & Tools • Adjacency Based Measurements • NetAct Optimiser 2.0 overview • Propagation Delay



Adjacency Based Measurements • Each cell has its own neighbouring cell list initially defined by radio Network Planning. This is a list of those neighbouring cells where the handover can be made. • The results of neighbour cell measurements can be used to optimise those lists. The benefits of optimised lists are better call quality and shorter handover delays. • To find strong candidates that are missing from actual definition • To locate and delete unused adjacencies • Identify and optimise badly performing adjacencies

RN C Serving BTS

UE s



Adjacency Based Measurements Counters M1013 Autodef SHO • M1013C0 Number of Intra Frequency SHO attempts • Counter is Updated when SRNC starts a Branch Addition or Branch Replacement procedure.

• M1013C1 Number of completed Intra Frequency SHO • Counter is updated when SRNC successfully ends the Branch Addition or Branch Replacement procedure.

M1014 Autodef IFHO • M1014C0 Number of Inter Frequency HHO attempts • Counter is updated when SRNC starts inter-frequency HHO

• M1014C1 Number of completed Inter Frequency HHO • Counter is updated when SRNC successfully ends inter-frequency HHO

M1015 Autodef ISHO • M1015C0 Number of Inter System HHO attempts • Counter is updated when SRNC starts inter-system HHO

• M1015C1 Number of completed Inter System HHO • Counter is update when SRNC receives RANAP:IU RELEASE COMMAND from core network after successful Inter System HHO



Adjacency Based Measurements Counters For each measurements (SHO, IFHO and ISHO) Statistic show: • # of HO attempts • # of HO completed (successful) to source and target cell objects Measurement is carried out in SRNC HO completion is considered successful if the SRNC during the handover decision does not detect any errors (errors in the source RNC side or failure messages from RRC/Iu/Iur/Iub interfaces) Object identifiers for M1013 and M1014 Source-RNC/Source-CID Target-RNC/Target-CID MCC/MNC Object identifiers for M1015 (ISHO) Source-RNC/Source-CID GSM-LAC/GSM-CID MCC/MNC For internal use 12 © Nokia Siemens Networks


SHO Share • HO Share provides distribution of HO attempts from the source cell • Useful detect neighbour relations which has exceptional amount of attempts

• It is possible to get the total number of outgoing HO attempts from the Autodef HO measurements by taking a sum over all the adjacencies reported for a source cell • SHO HO Share (M1013 AutoDef SHO) SHO _ Share _ RNC _ 903a

100 * sum( SHO _ ADJ _ INTRA _ FREQ _ SHO _ ATT ) =

Sum _ over _ all _ adja _ from _ the _ cell ( SHO _ ADJ _ INTRA _ FREQ _ SHO _ ATT )

• IFHO HO Share (M1014 AutoDef IFHO) IFHO _ Share _ RNC _ 904a

100 * sum( HHO _ ADJ _ INTER _ FREQ _ HHO _ ATT ) =

Sum _ over _ all _ adja _ from _ the _ cell ( HHO _ ADJ _ INTER _ FREQ _ HHO _ ATT )

• ISHO HO Share (M1015 AutoDef ISHO) ISHO _ Share _ RNC _ 905a


100 * sum( HO _ ADJ _ INTER _ SYS _ HHO _ ATT ) =

Sum _ over _ all _ adja _ from _ the _ cell ( HO _ ADJ _ INTER _ SYS _ HHO _ ATT )


HO Success per Adjacency • The HO success rate per adjacency can be calculated by using formulas below • Can be used to detect badly performing neighbours

• SHO Success per Adjacency (M1013 AutoDef SHO) SHO _ success _ per _ ADJS _ RNC _ 900a

100 * sum( SHO _ ADJ _ INTRA _ FREQ _ SHO _ COMPL ) =

sum( SHO _ ADJ _ INTRA _ FREQ _ SHO _ ATT )

• IFHO Success per Adjacency (M1014 AutoDef IFHO) IFHO _ success _ per _ ADJI _ RNC _ 901a

100 * sum( HHO _ ADJ _ INTER _ FREQ _ HHO _ COMP ) =

sum( HHO _ ADJ _ INTER _ FREQ _ HHO _ ATT )

• ISHO Success per Adjacency (M1015 AutoDef ISHO) ISHO _ success _ per _ ADJG _ RNC _ 902a



100 * sum( HO _ ADJ _ INTER _ SYS _ HHO _ COMPL) =

sum( HO _ ADJ _ INTER _ SYS _ HHO _ ATT )

Content • Neighbour optimization Methods & Tools • Adjacency Based Measurements • NetAct Optimiser 2.0 overview • Creating ADJx based on PM data

• Propagation Delay



Automated Adjacency Optimisation for 3G in Optimizer 2.0 • Accurate and efficient process for optimising operational WCDMA cell adjacencies

• Measurement based optimisation • Current adjacency status analysis • Deletion of unused adjacencies based on KPIs • HO attempts, HO success

• Adjacency candidate identification, activation and measurement • Interfering intra-frequency cells • Cell pair Ec/No difference from WCDMA • Neighbour cell signal strength from GSM

• Final adjacency list optimisation • Scrambling code re-allocation

• Full visibility and control to the user



Creating new adjacencies • A fast way to identify missing intra-frequency adjacencies • Interference measurements colleted from RNC • New adjacencies can be created based on that statistics

• Rotation method used to achieve the optimal lists for other adjacency types • Optimizer creates adjacency candidates • Candidates are downloaded to network and measured • Statistics collected directly from RNC • Cell pair Ec/No difference • Successful BSIC verifications & BSIC verification time

• Final adjacency list is generated



Content • Neighbour optimization Methods & Tools • Adjacency Based Measurements • NetAct Optimiser 2.0 overview • Creating ADJx based on PM data (AutoDef) • Creating ADJx based on DSR measurements (ICSU) • WCDMA Interference analysis




Creating ADJx based on PM data (AutoDef) • Optimizer will show adjacency based SHO amounts for undefined neighbours • Purpose is to search all ADJS and ADJG new neighbours which are within certain max distance • Example 1-5 km in urban area and 4-10 km outside urban area.

• After that only those will be selected which have enough SHO/ISHO attempts. • The selected neighbours could be provisioned straight away to the network



How to create Missing ADJx based on PM data-1 1. Select area from the map And start the ADJ Optimization tool

2. Select ADJG, ADJS and ADJW types



How to create Missing ADJx based on PM data-2 3. Select right actions from rules, common Deletion and Creation tabs



How to create Missing ADJx based on PM data-3 6. Save plan from here with any name

5. Start from here

4. Purpose is to search all ADJS and ADJG new neighbours which are within certain max distance like 1-5 km in urban area and 4-10 km outside urban area. After that only those will be selected which have enough SHO/ISHO attempts. For internal use 22 © Nokia Siemens Networks


How to create Missing ADJx based on PM data-4

7. List all new neighbors



How to create Missing ADJx based on PM data-5 8. Select the whole week or one day for PM data analysis

9. Select the right profile to browser (ADJG, ADJS)

10. Update the list of Neighbours from here 11. Sort according to the PM attempts



How to create Missing ADJx based on PM data-6 12. See the ADJ on top of the map

13. Provision the selected neighbors to the network

Note ! These neighbors are defined only for one way direction. See next slides how to make those bidirectionally (Refresh actual operation with RAC)



How to create Missing ADJx based on PM data-7 14. Open the CM data exchange under the main window 15. Select refresh actual and wait Until the data is updated

16. Open the adjacency optimization without selecting any tabs from Deletion or Creation, just to find just created one way ADJx



How to create Missing ADJx based on PM data-8 17. Save the plan and list the planned elements 18. You can see now the ADJx neigbours which can now provisioned to the network



Content • Neighbour optimization Methods & Tools • Adjacency Based Measurements • NetAct Optimiser 2.0 overview • Creating ADJx based on PM data (AutoDef) • Creating ADJx based on DSR measurements (ICSU) • WCDMA Interference Analysis




Creating ADJx based on DSR measurements (ICSU) • Detected set measurements coming from undefined neighbours (based on ICSU logs) • Aim is to find source of interference • cell having many DSR results but no SHO attempts (with neighhbour list combination list)

• Solutions • Add found cell to the neighbour • Downtilt to decrease the interference

• DSR measurements are suitable also for ADJG neighbours



DSR activation • When DSR is not activated, UE monitors only cells in its NCL (either read from BCCH or sent from RNC in SHO case). • When DSR IS activated, UE scans ALL scrambling codes in same frequency band and if cells are found that fulfil certain criteria, UE reports this/these cell(s) as detected cells. • criteria for detection is that UE has to be able to detect if Ec/N0 is greater than -18 dB • for a DSR to be triggered, detected cell/s must fulfil "normal" HO criteria, i.e. for example, are within the reported range relative to P-CPICH of strongest AS cell. • Details of activation :MML command that is sent to RNC that sets some flag active and RNC orders UE to measure and report. It can be done by HIT macro, but Optimizer is not (supposed to) using them but same commands that are in HIT macros are sent directly to RNC.



1. Select scope, area from map Or individual site or cell

2. Select tools/ Adjacency Optimization 3. Select ADJS,(ADJS=3G/3G, ADJG=3G->2G,ADJW=2G->3G ADJI= 3G-3G,ADJE=2G-2G)



4. Rules: Fill and keep adjacencied in same Site and Fill bidirectionality 5. Common: Use Bidirectional Optimisation and Enable Changes between Scope and Buffer area 6. Change DSR parameters from here



7. If you want look at cells,which are seen by DSR, but are not in combined list (no SHO attempts to that one), select SHO attempts big enough to find The cells which are purely interference sources



8. Start optimization and you see the found created adjacencies Tästä halutut solut kartalle in browser, if you want to see the cells on map, the plan should be stored



9. Here you can see DSR reports

10.If you want implement this neighbor, just select both directions from Adjacencies tab: list to browser and further by selecting both directions and with right mouse button: Provision instantly For internal use 35 © Nokia Siemens Networks


Select both directions and provision instantly



Optimizer does not create neighbours if one rule below is true.However if all missing neighbours want’s to be seen, these rules can be omitted (and see all interfering cells based on DSR). If you want to add neighbors, it is recommended to check SC problems with by Scrambling code allocation tool. Don’t use this in normal operation.



11. It is possible to find ADJG/ADJW based on Measurement data, if bidirectional Neighbours are wanted both ADJG and ADJW should be selected







WCDMA Interference Analysis •

it is possible to see WCDMA internal interference situation of certain cell which is caused by other WCDMA cells (in terms of distance,RSCP and EcNo level). Also non-neighbours can be measured with DSR.

•

Both Incoming and Ongoing interference levels can be studied with certain cell pairs.

•

Interference information is based on UE measurements where the signal strength and quality of every Primary Scrambling code is reported to RNC.

•

Before interference analysis the following actions should be done 1. Start WCDMA interference measurements 2. Activate measurements from Interference matrix (from selective RNCs or all measurements) 3. Start Interference analyzer for WCDMA –tool (from Scope, RNC or site) 4. Analyze the selection which was done in step 3



WCDMA Interference Analysis • The analysis shows for example the number of measured adjacencies with number of reports (SHO + DSR measurements) and if it is neighbour or not. • If there is lot of reports from non-neighbour cell it would make sense to add it to the neighbour, at least if the distance is reasonable and if the RSCP levels are high. This will mean that the cell could be interferer, especially if there is not much SHOs (low SHO share %) to that cell (even with SHO combination). DSR result from no neighbour

RSCP,EcNo criteria







Propagation delay counters • PRACH propagation delay statistics is presented using a distribution consisting of 21 counters M1006C128-M1006C148. One of the counters is updated by value 1 when the UE sends RRC Connection Request or Cell Update. • Each counter covers one or more PROP_DELAY values and the mapping of measured values to counters can be controlled by WCEL parameter PRACHDelayRange that defines five different mapping tables for various cell sizes

Range 60 km (this is fixed in RAS06) bin from(m) PROP_DELAY (from) to(m) PROP_DELAY (to)

0 1 0 234 0 1 234 468 0 1

2 3 4 5 6 7 8 468 936 1170 1638 2106 3042 3978 2 4 5 7 9 13 17 936 1170 1638 2106 3042 3978 4914 3 4 6 8 12 16 20

9 4914 21 6084 25

10 6084 26 7020 29

bin size(m)

234 234

468

1170

936


234

468


468

936

936

936

11 12 13 14 15 16 7020 7956 10062 14976 19890 25038 30 34 43 64 85 107 7956 10062 14976 19890 25038 29952 33 42 63 84 106 127 936

2106

4914

4914

5148

4914

17 29952 128 34866 148

18 34866 149 40014 170

4914

5148

19 20 40014 50076 171 214 50076 infinite 213 10062

PRACH example, RNC, 2 weeks data Main Distance 468-936 m 600000

120.00%

500000

100.00%

400000

80.00%

300000

60.00%

200000

40.00%

100000

20.00%

0

0.00%

0 1 2 3 4 5 6 7 1 8 1 9 2 0 0 1 2 3 4 5 6 7 8 9 _ S 1 _ S 1 _ S 1 _ S 1 _ S 1 _ S 1 _ S 1 _ S_ S_ S_ S_ S S_ S S_ S S_ S S_ S S_ S S_ S S_ S S_ S S_ S 1 S A L A L A L A L A L A L A L A L A L A A S A S A S A S A S A S A S A S A S A S A S L L L C L L C L L C L L C L L C L C _ Y C _ Y C _ Y C _ Y C _ Y C _ Y C _ Y C _ Y C _ Y C _ C _ Y C _ Y_ Y C _ Y_ Y C _ Y_ Y C _ Y_ Y C _ Y_ Y Y L A E L A E L A E L A E L A E L A E L A E L A E L A E L A L A L A L A L A L A L A L A L A L A L A L A E E D E D E D E D E D E D E D E D E D E D E D D D D D D D D D D _ H_ H_ H_ H_ H_ H_ H_ H_ H_ D _ _ _ _ _ _ _ _ _ _ _ H H C H C H C H C H C H C H C H C H C H C H C A C A C A C A C A C A C A C A C A C C R A P R P R P R P R P R P R P R P R P R R A R A R A R A R A R A R A R A R A R A R A P P P P P P P P P P P P For internal use 44 © Nokia Siemens Networks


average CDF

PRACH example, WCEL, one day Class 14: 14.5 km-20km 450

400

350

300

Sorted by Class12

PRACH_DELAY_CLASS_12 PRACH_DELAY_CLASS_13 PRACH_DELAY_CLASS_14

250


200


150

100

50

0

2 5 8 9 3 0 1 0 9 3 8 9 9 6 6 9 5 9 3 1 6 8 8 0 3 0 0 8 1 3 8 1 6 8 3 9 1 9 7 7 6 5 7 4 9 4 7 6 2 4 0 5 6 1 2 6 1 2 5 7 2 1 5 8 6 0 5 7 6 0 1 9 6 0 6 1 8 3 5 8 2 2 0 1 3 0 6 5 6 0 6 0 5 7 5 9 8 6 8 4 7 4 1 5 2 0 0 9 6 4 4 4 4 1 4 4 4 4 1 4 4 4 4 1 1 1 4 4 4 4 4 4 4 4 4 6 4 5 4 6



Neighbour Optimization RU10_Nokia

Recommend Documents