UMTS Introduction of Network Coverage Evaluation Based On MR0309

Security Level:
Introduction of Network Coverage

Evaluation Based on MR
www.huawei.com
HUAWEI TECHNOLOGIES CO., LTD.
Contents
MR based Geolocation Principle Introduction
MR Basic and Location principle
Traffic map value
Coverage Evaluation Introduction

AGPS MR Collection
Coverage Evaluation based on AGPS MR
Case Introduction
Page 2
MR Traffic Map Solution Overview

Accurate Coverage Visualization
Coverage Problems Auto Identification
Geographical visualization of
RSCPECIOUE Power and Cell
Edge based on real user
information
With configurable thresholds

and parameters, auto identify
spots with poor coverage, pilot
pollution and overshooting
problems
Security Level:
Smart Site Planning

New site address is auto located
according to weak coverage spots
and users distribution
Spot with Weak Coverage
www.huawei.com
Service Product Status: Available for R12, R13. Tool: OMStar

Page 3
Traffic Map Introduction - MR Description

Measurement Report
Measurement Control
Measurement Report
Periodic Trigger
(MR Module)
Event Trigger
(HO Module)
MR MAP :
According to corresponding
locating algorithm, make the
information brought by UE
(such as RSCP, Ec/Io, TX
Power and so on) display on
the map.
Page 4
MR Location Calculation Process
Feature Database
MR Data
Calibrated Engineering Parameter
Calculated Location
Point
Location Calculation
Time
9/16/2010 13:29:13.600
9/16/2010 13:29:23.600
IMSI
440001090253516
440001090253516
RNC
Cell
2305 23405
2305 23405
Longitude
140.116544
140.117095
Latitude
36.12388
36.124334
Nastar
MR Location Calculation Process

Page 5
Traffic Map Introduction - MR Location principle

Locating PrincipleCELLID+RTTFeature matching MethodAGPS
Signal Features
Moving Unit
1.CellID + RTT
2. Feature matching Method
3.AGPS
From RTT measurement results,

calculates the distance from the UE
to serving cell, 1 RTT result
determines 1 circle, 3 results
determine 3 circles, and the cross
points of the 3 circles are the real
location.
Use the methods of DT,AGPS and

coverage simulation to establish
network feature database, such as
RSCP, Path info, Time Delay etc.
And then compare MR with the
feature database. Ue location is
where the bin match best.
RNC collect information related

to GPS satellites, and the data
is given to the UE needed for
positioning, UE search to GPS
satellite signals and estimate
location.
The SAS is a separate device, which

is connected to an RNC through
the Iupc interface.
Page 6
The SAS is used for UE positioning
Baseline of MR Location Accuracy

Parameters
Input / Output
U / SU
Output
80 120m
Site distance
Input
300 800m
Minimum MR message collection on

each position
Input
5000 MR Messages
Site Lat / Long tolerance
Input
< 15m
Antenna azimuth tolerance (Max)
Input
< 10 degree
Drive test data support (Cover major

road)
Input
Yes
Digital map resolution
Input
5m (with building information)
Traffic map accuracy

*note 1
Dependences
*Note 1: The output accuracy is subject to the input requirement as specified.

Digital map resolution>50mthe impact for accuracy is about 20%~30%.
No drive test :
for modified feature database, the impact for accuracy is about 30%.
for propagation model ,the impact for accuracy is about 30%.
Scenario not applicable to:

Single sites only
Site on hill /Sea / Lake
Page 7
DT/CQT MR Accuracy
DT Scenario
CQT Scenario
Cell Quality Test
Scenario
Accuracy
CQT
14m @ 67%
DT
34m @ 67%
Page 8
MR Location Accuracy by Using AGPS to Correct Character DB
Character DB
Accuracy
Before Correction
626m @ 67%
After Character DB
153m @ 67%
Correction Using AGPS Data
Using AGPS Data to
87m @ 67%
Forming Character DB
Specification:
Before Correction: Using simulation method by propagation model to
make the character DB WITHOUT any data correction.
After Character DB Correction Using AGPS Data: Based on the
prediction character DB, Using the AGPS data to correct the prediction
character DB.
Using AGPS Data to Forming Character DB: Directly make the
character DB by AGPS data WIHOUT prediction.
Page 9
Contents
Traffic map value

AGPS MR Collection
Case Introduction
Page 10
Traffic Map Analysis Value

1.Real user RSCP coverage
2.No drive test Required
1.Real user Ec/Io coverage

2.No drive test Required
Ec/Io Distribution
RSCP Distribution
1.Real user serving cell distribution

2.More precise than simulation result
Serving Cell Edge Distribution

1.Overshooting Analysis
2.Handover Analysis
3.Interference Analysis
4.Traffic Unbalance Analysis
Remark: Max MR records cell as serving cell per lattice.
Page 11
Traffic Map Analysis Value

Handover Ratio Distribution
Call Drop Rate Distribution
Type UE Distribution(Example for iphone)
Real user distribution/High User Spot Distribution
Page 12
Contents
Traffic map value

AGPS MR Collection
Case Introduction
Page 13
Coverage Assessment Workflow - General
005.005.001
005.005.002
005.005.003
Activate MR
period Reporting
Activate AGPS
Reporting
(Optional)
KPI Monitoring
Field Engineer
LMT
Coverage Performance Assessment
Nastar/Mapinfo/ArcGIS
005.005.006
005.005.012
005.005.010
Output Topology
Evaluation Report
Output Coverage
Evaluation Report
005.005.0011
GIS Display
005.005.007
Coverage Audit
Cell/RNC/LAC
Edge Audit
005.005.009
Hotspot Audit
005.005.008
Omstar
Server/Nastar
V6R10
Interference Audit
002
005.005.005
Collect the data
Geo-Location
Processing
Raw MR data (7
days)
U-net
003.004
Probe Assistant
Engineering Parameter
Digital MAP
Generate the
Feature Database
005.005.004
DT Data
Adjust
Propagation
Model (Optional)
Page 14
Coverage Assessment Workflow Based on Prediction
Page 15
Coverage Assessment Workflow Based on AGPS

Specification:
Emphasis on how to collect
AGPS MR data
Other steps can be referred in instructions.
Mapping method is helpful

to give rise and investigate to
obvious RF problems
Page 16
How to collect the AGPS MR data

Standalone mode:
DO NOT need GPS hardware support,
UE can search satellite to locate.
UE Based mode: Network sends
the satellite data to UE, and help UE to
search the satellite, then UE computes
the location:
GPS hardware can be integrated in the
GCGa board, and also can be independent
by using GCUa + GPS.
Generally, One RNC configures one GPS.
(1) AGPS function is controlled by license, so we need confirm whether the license can
be valid : LST LICENSE:FN=*********.dat;
(2) RNC Version R12 and above R12
Page 17
MML Command for AGPS MR Collection

RNC
SET UMRCTRL: MRTRAFFICTYPE=TRAFFIC_AMR-1&TRAFFIC_VP-1&TRAFFIC_BE-1&TRAFFIC_OTHER-1,
MRINTRAFREQPERIOD=D12, MRRPRTCFGSWITCH=MR_PRD_INTRA_FREQ_MEAS_SWITCH_11&MR_EVT_INTRA_FREQ_MEAS_SWITCH_2-1&MR_PRD_INTER_FREQ_MEAS_SWITCH_31&MR_EVT_INTER_FREQ_MEAS_SWITCH_4-1&MR_PRD_INTER_RAT_MEAS_SWITCH_51&MR_EVT_INTER_RAT_MEAS_SWITCH_6-1&MR_DL_BLER_MEAS_SWITCH_70&MR_UE_TX_POWER_MEAS_SWITCH_8-0&MR_UL_SIR_MEAS_SWITCH_90&MR_LCS_MEAS_SWITCH_10-1&MR_RACH_MEAS_SWITCH_11-0,
MRLCSMETHODSWITCH=LCS_AGPS_UEBASED_SWITCH_1-1&LCS_AGPS_STANDALONE_SWITCH_20&LCS_CELLID_RTT_SWITCH_3-1, MRDLBLERPERIOD=D12, MRUETPPERIOD=D12, MRULSIRPERIOD=D12,
MRLCSPERIOD=D12;
SET USMLC:SMLCMETHOD=CELLID_CENTER-1&CELLID_RTT-1&OTDOA-1&AGPS-1,
AGPSMETHODTYPE=UE_BASED, AGPSADDASSDATASENDFLAG=DELIVERY,
UEASSAGPSASSDATASWITCH=REFERENCE_TIME_FOR_GPS-1&REFERENCE_LOCATION_FOR_GPS1&DGPS_CORRECT-0&GPS_NAVGMODEL-1&GPS_IONOSPHERIC_MODEL-0&GPS_UTC_MODEL0&GPS_ALMANAC-0&GPS_ACQ_ASSIST-1&GPS_REALTIME_INTEGRITY-0,
UEBASAGPSASSDATASWITCH=REFERENCE_TIME_FOR_GPS-1&REFERENCE_LOCATION_FOR_GPS1&DGPS_CORRECT-1&GPS_NAVGMODEL-0&GPS_IONOSPHERIC_MODEL-1&GPS_UTC_MODEL0&GPS_ALMANAC-0&GPS_ACQ_ASSIST-0&GPS_REALTIME_INTEGRITY-1, MAXGPSSATS=16,
OTDOAMETHODTYPE=UE_ASSISTED, CELLIDRTTMETHODTYPE=UE_ASSISTED, EMERGLCSSWITCH=OFF,
OPTIONALIESENDSWITCH=R5_POSDATA_SWITCH-0&LAST_KNOWN_SERVICE_AREA_SWITCH-0,
FORCEDSHOSWITCH=OFF, LCSWORKMODE=RNC_CENTRIC;
Page 18
MML Command for AGPS MR Collection

Cell:
Set MR collection scope:
SET UMRSCOPE:CELLID = 37851, SCOPETYPE=ByCellId, MRSCOPECTRL=ON;
SET UMRSCOPE:CELLID = 37852, SCOPETYPE=ByCellId, MRSCOPECTRL=ON;
If need switch MR collection off, just transfer MRSCOPECTRL=ON TO MRSCOPECTRL=OFF
ADD USMLCCELL information:

ADD USMLCCELL: RNCID = 2340, CELLID = 48143, CELLLOCCFGTYPE = CELL_ANTENNA, GCDF
= DEG, MTRLGY= MET, ANTENNALATITUDEDEGREE = 29883800, ANTENNALONGITUDEDEGREE
= 121870500, ANTENNAALTITUDEMETER =40, MAXANTENNARANGE=1000,
ANTENNAORIENTATION=2700, ANTENNAOPENING= 1200, CELLIDRTTACTIVATEFLAG =
INACTIVE, OTDOAACTIVATEFLAG = INACTIVE, AGPSACTIVATEFLAG = ACTIVE;
When MR collection finished, we need rollback the switch above.

Accessories for tools of ADD USMLCCELL AND rollback it:
Page 19
Impact for Network after AGPS Switch ON

Impact for SPUa Load
Smart Phone penetration rate SPUa load enhanced
40%
5%
1 week collection
Impact for Connection Performance

Impact for Keeping Performance
Impact for Handover Performance
Command
CS RAB Connect
successful rate
AGPS ON
98.10%
AGPS OFF
99.83%
Call drop rate
AGPS report rate
CS: 0.478%
PS: 0.76%
CS: 0.42%
PS: 0.42%
4.5%~5%
Note: If need to degrade the impact of

the connect performance further, we can
switch AGPS on in batches.
Page 20
Why Impact Network after AGPS Switch ON

When switch AGPS on, RNC will send measurement control including
mass of satellite data to UE.
Sometimes the UE cannot send the response signal to RNC in time, and
then cause the Iu release.
In order to solve this problem, we attempt several methods, One effective
method is to close the GPS_NAVGMODEL in SET USMLC command.
Because it can reduce the num of satellite data from 740 bytes to 130 bytes.
Negative impacts cannot be avoided (just reduced)
To limit the impact, we can switch AGPS in batches.
Page 21
Contents
Traffic map value

AGPS MR Collection
Case Introduction
Page
Page22
22
Create Character DB
Prediction
Use DT to
correct
prediction DB
Based on
AGPS
Dependant on propagation model and engineer

parameters. Accuracy is low.
Use DT data to correct the prediction DB, But just
correct the DB on road. Character DB out of road
will depend on prediction.
Use AGPS MR to create character DB, not
dependant on the engineer parameters and
propagation model. Accuracy is highest.
Page 23
Network Evaluation based on AGPS MR

Map Analyzer
Cell List
RSCP
Ec/Io
Pilot pollution
Best Sever
Traffic
LAC distribution
Event distribution
Terminal distribution
Overshooting Cell list

Cell list of unreasonable LAC distribution
Cell list of unbalance for UL and DL
Cell list with Fault Antenna
Cell list of unreasonable overlapping zone
Cell list of pilot pollution
Cell list of abnormal coverage area
Page 24
Network Evaluation
RSCP Coverage
For Urban
Ec/Io Coverage
For Urban
We also can display RSCP, Ec/Io for each carrier to find the poor coverage
area on the map.
Page 25
Network Evaluation
Pilot Pollution
Best Sever
From the pilot pollution map, get the more serious pilot pollution area.
From the Best server map, get the cell coverage range.
With this two map analyze which cell cause the pilot pollution
Page 26
Network Evaluation
AMR Traffic Distribution
PS Traffic Distribution
Use RAB field to distinguish the service type, RAB = 1, means AMR, RAB = 4,
means BE.
From the pictures above, we can find the hotspot and prior to pay attention to the
hotspot, to suggest network capacity expansion.
Page 27
Network Evaluation
LAC Distribution
LAC Distribution
From the LAC distribution and the traffic distribution, we can judge Whether the
LAC boundary is reasonable or not. LAC boundary cannot be divided at the dense
traffic area.
Page 28
Network Evaluation
Call Drop Event Distribution
iPhone Hotspot
Associating MR and PCHR logs, we can display the event map, such as call drop
map, HO failure map, etc.
We also can display the coverage map and event map of some type terminals such
as iPhone.
Page 29
Network Evaluation - Overshooting

According to the relationship of distance and MRs to judge whether the cell is
overshooting or not.
From the two pictures, we can clearly see the overshooting cells by Best Server
map and MRs PDF.
All the overshooting cells can be given by Nastar tool.
Page 30
Network Evaluation Faulty Antenna

Rules:
(1) Choose periodic MRs with AGPS.
(2) Compute the angle of every AGPS MRs to Best Server azimuth
(3) Stat the ratio = the angle > 90 degree/ all periodic MRs
(4) If the ratio >50% and there are above two cells with different azimuth in one
site
NodeBID
Cellid
Faulty specification
3158
31582
123 sector fault
3158
31583
123 sector fault
3158
31581
123 sector fault
3761
37616
23 sector fault
3761
37613
23 sector fault
3761
37612
23 sector fault
3761
37615
23 sector fault
3746
37464
12 sector fault
3746
37461
12 sector fault
3746
37462
12 sector fault
Page 31
Network Evaluation Overlapping Zone

Rules:
(1) Choose periodic MRs.
(2) Stat the overlapping zone as couple cells
(3) Compute the ratio = MRs of all Non-Best server and best server are in
active set / all MRs in best server.
Cell1
Cell2
35351
37281
31391
38781
40411
37511
26961
37352
35353
37283
31393
38783
40412
37613
26962
37353
MRs of all cell1

Overlapping
Overlapping
and cell2 in active MRs inCell1
MRs inCell2
degree1
degree2
set
3960
4082
97.01%
47351
8.36%
1281
1337
95.81%
21195
6.04%
2500
2976
84.01%
78696
3.18%
1952
2574
75.84%
20303
9.61%
10695
14633
73.09%
17296
61.84%
2136
3128
68.29%
23296
9.17%
3270
4820
67.84%
36321
9.00%
14433
22436
64.33%
32855
43.93%
We can use it to check the overlapping zone of the cells , reduce the SHO rate,
and remove the redundant cells.
Page 32
Network Evaluation Pilot Pollution

Rules:
(2) Neighbor cells Num > 3 and top 4 cells RSCP > -95dBm and Ec/Io < -12dB
(3) 1st Cells RSCP - 4th cells RSCP < 5dB
Pilot pollution subject is included in Nastar V6R9 or above. So we can use

Nastar to analyze pilot pollution.
From the result, we can see the polluted and polluter cell for CS/PS/Total.
Page 33
Network Evaluation LAC distribution

Rules:
(2) Stat the MRs that includes the cells belonging to different lac in active set.
RNCID
CellID
MRs in different LAC
2354
31642
612
2354
31711
788
2354
31613
847
2354
20833
745
2354
37592
618
2354
31434
408
2354
40793
856
2354
31111
153
2354
37414
1523
2354
31492
111
2354
31046
512
2354
37535
785
2354
37124
1161
The cells with more MRs in different LAC

is belonging to unreasonable LAC
distribution, we should change the cells
coverage or the cells lac ownership.
Page 34
Contents
Traffic map value

AGPS MR Collection
Case Introduction
Page
Page35
35
Case 1 introduction: Weak Coverage Optimization

Few users Area
Bad RSCP
Azimuth from 240 to 200
Azimuth from 130 to 170
How to improve?
Tuning Summary
High traffic(User num)
Point
Before
Best
After
RSCP(dBm)
Gain(dB)
A
-97.5
-91
6.5
B
-92.5
-92
0.5
Page 36
C
-107.5
-102.5
5
D
-87.5
-84
3.5
E
-105
-93
12
Case 2 introduction: Automatic New Added Site Based on Weak Coverage
Automatic new added site
New site address is auto located where RSCP and users distribution meet specified
condition in specified amount of consecutive grids.
Automatic site planning: site address
Poor coverage area
Page 37
Case 3 introduction: Pilot Pollution and Overshooting Problem
Pilot Pollution Problem : Serving cell and Polluted cell

Pilot Pollution Analysis
Main
Pollution
Suggestion
Serving
Cells
21906
Control the
21506
13106
coverage of
25006
pollution cells
6906
Overshooting Problem : overshooting distribution

Lon
Lat
113.6341 34.7821
Azimuth RNCID CELLID

150
768
21952
Page 38
Overshooting
Distance(m)
117
Overshooting
Area(sq.m)
1666.0309
Case 4: Hot Traffic Spot Finding (1/2)

PS Users Num Distribution
PS EcIo Distribution
PS hot spot: MR larger than some threshold
PS hot spot: MR VS Performance
11% top cells, occupy 27% PS traffic
Page 39
Case 4: Hot Traffic Spot Finding (2/2)

Precision Verification
Geography
Magnification
Results
Verification
Economic
Comparison
Before
(with traditional way )
After
(with traffic map)
Analysis area Human resource(per day)
5km*5km
site num:70
Gain
Time (day)
Hot Spot num
Unsure
14
60%
71%
Page 40
Page 40
Case 5 introduction: Real Coverage Distribution (No Location)
No Location MR: only MR data, Engineering Parameter, RNC Configuration Data (No Feature Database)
Page 41
Page 41
Thank you
www.huawei.com
Copyright2011 Huawei Technologies Co., Ltd. All Rights Reserved.
The information in this document may contain predictive statements including, without limitation, statements
regarding the future financial and operating results, future product portfolio, new technology, etc. There are a
number of factors that could cause actual results and developments to differ materially from those expressed or
implied in the predictive statements. Therefore, such information is provided for reference purpose only and
constitutes neither an offer nor an acceptance. Huawei may change the information at any time without notice.
1. Question: current accuracy can not satisfy customers requirement. And traffic map dont support indoor
coverage so far.
Answer: 1) locating accuracy is related with the actual geography environment for example site distance, the
short distance, the higher accuracy. Modify feature database (digital map 5m is best and drive test is needed) can
improve the accuracy. Another way open AGPS report, as long as UE support, the accuracy also can be improved.
2. Question: current traffic map dont support PS throughput geography display.
Answer: The roadmap for PS throughput display will be in R14 at 2012Q2; So far we can combine user
number/MR number etc conditions to indirectly recognize.
3. Question: how about the impact for RNC CPU load after open the MR switches.
Answer: In lab test based on Canada traffic model, after open MR switches, the SPU CPU Load will increase 10%
relatively(50%55%). Another side, MR switches support NodeBs and Cell level. It can be opened several
NodeBs or cells switches. After collection the MR data (normally one week), then close the switches.
Page 43

UMTS Introduction of Network Coverage Evaluation Based On MR0309

Uploaded by

Copyright:

Available Formats

UMTS Introduction of Network Coverage Evaluation Based On MR0309

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

UMTS Introduction of Network Coverage Evaluation Based On MR0309

Uploaded by

Copyright:

Available Formats

Security Level:

Introduction of Network Coverage

HUAWEI TECHNOLOGIES CO., LTD.

Traffic map value

Coverage Evaluation Introduction

MR Traffic Map Solution Overview

Coverage Problems Auto Identification

With configurable thresholds

Smart Site Planning

Spot with Weak Coverage

Service Product Status: Available for R12, R13. Tool: OMStar

Traffic Map Introduction - MR Description

MR Location Calculation Process

Calibrated Engineering Parameter

MR Location Calculation Process

Traffic Map Introduction - MR Location principle

2. Feature matching Method

From RTT measurement results,

Use the methods of DT,AGPS and

RNC collect information related

The SAS is a separate device, which

HUAWEI TECHNOLOGIES CO., LTD.

Baseline of MR Location Accuracy

Minimum MR message collection on

Site Lat / Long tolerance

Antenna azimuth tolerance (Max)

Drive test data support (Cover major

Digital map resolution

5m (with building information)

Traffic map accuracy

*Note 1: The output accuracy is subject to the input requirement as specified.

Scenario not applicable to:

HUAWEI TECHNOLOGIES CO., LTD.

MR Location Accuracy by Using AGPS to Correct Character DB

Correction Using AGPS Data

Using AGPS Data to

Traffic map value

Coverage Evaluation Introduction

HUAWEI TECHNOLOGIES CO., LTD.

Traffic Map Analysis Value

1.Real user Ec/Io coverage

1.Real user serving cell distribution

Serving Cell Edge Distribution

Remark: Max MR records cell as serving cell per lattice.

HUAWEI TECHNOLOGIES CO., LTD.

Traffic Map Analysis Value

Call Drop Rate Distribution

Type UE Distribution(Example for iphone)

Real user distribution/High User Spot Distribution

HUAWEI TECHNOLOGIES CO., LTD.

Traffic map value

Coverage Evaluation Introduction

HUAWEI TECHNOLOGIES CO., LTD.

Coverage Assessment Workflow - General

Coverage Performance Assessment

Collect the data

HUAWEI TECHNOLOGIES CO., LTD.

Coverage Assessment Workflow Based on Prediction

HUAWEI TECHNOLOGIES CO., LTD.