07_Nokia 3D Geo-Location

NPO Services 3-D Geolocation: Performance monitoring, Planning & Optimization Nazir Uddin SSC AMEA NPO 17th Sept 2014

1 17/09/2014 © Nokia 2014 - Geo – location - NPO Confidential

Nokia services for the best performing networks & mobile broadband experience Modules QoS Differentiation Consultancy

Give each individual user the experience they want

CEMoD Driven Optimization

Experience optimization

SFNO VoLTE Quality Analysis & Optimization

LTE MBB Analysis & Optimization

Application Impact Management

3-D GL based Advanced Optimization

IPv4 to IPv6 Migration Consultancy

E2E Mobile Backhaul Evolution

Improve the performance of services for a better QoE

Service Network Optimization Optimization Technology evolution Upgrade to new technology when and where it is really needed

2 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

Minimize OPEX & CAPEX, while creating better performance

Public announcement Link to press release

• Nokia Networks […] bought a broad solution for advanced geolocation capabilities from NICE Systems (NASDAQ: NICE) to enhance the planning and optimization of mobile networks • Nokia Networks gains access to tools, technical expertise, and the right to further develop these capabilities 3 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

3D Geolocation – adding the 3rd dimension

Because the World is not flat Introducing 3D monitoring …

Best in class accuracy - horizontally and vertically

Drill down from network map to any single event

Mass geo-location based RF optimization

3D hot-spot location for small cell planning



Continuous 3D Geo-Location monitoring

60+m

 48-60m  36-48m  24-36m  12-24m 

0-12m

All

… for unparalleled accuracy of network performance 4 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

Traditional Methods vs Geolocation Weighted “Land Use & Building Use” Map Poor accuracy since based on assumptions.

OMC Cell Statistic and Propagated Best Server Coverage Accuracy highly dependent on cell size.

2D Geolocation using billions of MRs from Real Subscribers.

3D Geolocation using billions of MRs from Real Subscribers.

Accuracy dependent on algorithm used. Hybrid algorithm usually used.

Best in class for accuracy which separates inbuilding floor by floor, and outdoors.


3D RSS Fingerprinting & Adaptive Learning Algorithm for Geolocation

1

Ray Tracing •

• • • •

2

Receive Strength Signal - Transmitter power - Antenna gain - Path loss Free Space Fresnel Zone Multiple Knife Edge Proprietary prediction model: HEX3D B

Power Measurements

3D Geolocation • •

Create 3D prediction grid For each MR -

-

•

Calculate zone boundaries from TA derived distance using geometry Calculate matching factor to 3D prediction grid

Adaptive learning algorithm selects exact location

Hyperbola

Power prediction grid

B

Match to prediction grid

Cell

RSCP

A (PD 1)

-70

B

-75

C

-98

Highest matching score

C

System Frame Number

A


Distance ratio

Cell

Tm, OFF

A

56, 216

C

2261, 76

In-Building density maps • • •

•

Locate multiple MR and events for each call Observe speed and cell changes during the call Trace to building and calculate event density - Set performance threshold ( e.g. EcNo -14 dB) - Extract number and proportion of samples above and below threshold Produce performance density maps

3

High Level Architecture Events/MSR

(GPEH)

(Megamon)

Soft probes Note: Support of H// PCHR (UMTS) planned in roadmap

Adaptor & Parser

RF Events Database

Location Engine


Located RF Events

Monitoring

Optimization Engine

Expansion Planning

3D Geolocation Capabilities • RF Events Positioning: - Outdoors and indoors, in-building and at different floors of buildings

• Types of Maps: - Area (bins), roads, buildings and floors (height above ground) - Filtering options: indoor/outdoor, static/in-motion, per IMSI • 3D Parameter Maps: - Ec, Ec/Io, Polluting sectors, Active set links, UE Tx power - Voice traffic, Data traffic, HS traffic, User traffic • 3D Event Maps: Voice, Data, Move to GSM, Drops, Polluted • Drill down capabilities: - Date, Time, RNC ID, IMSI, Reference Cell, PD, location (X,Y,Z) - For each measured cell: RSCP, Ec/Io, DL/UL service type, DL/UL throughput, Drop reason


3D Geolocation Traffic Hotspots Location


Unparalleled Accuracy In Locating Traffic Hotspots Identify network traffic hotspots, both indoor (at different floors) and outdoor where offload solutions can be deployed.

Traditional 2D : 50m bin

Slice: 12-24m

Load

vs

Very high High Medium Reasonable Low 2D Hotspot location

Purpose 10 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

3D: Individual buildings

3D Hotspot location

Hotspots Detection #1

Hotspot accurately detected with abusive users

Data traffic (Mb) - 50m bin Analysis

Top 15 IMSIs distribution Dominant IMSI - Internet Cafe 76.4%

6.4%5.8%2.6%2.2%1.8%1.0%0.9%0.6%0.5%0.4%0.2%0.1%0.1%0.1%

Building analysis 11 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

Hotspots Detection #2 with single cell time-based analysis 5 days aggregation

Bin analysis

Bus Station

College


Sun - 12:00 - 13:00

Sun - 08:00 - 09:00

Hotspots Detection #2 with single cell time-based analysis

HsDPA Traffic (Gb)

30

Monday

25 20 15

Sunday

10 5 0

7

8

9

10


11

12

13

14

15

16

17

18

19

20

21

22

23

Sun - 20:00 - 21:00

Mon - 08:00 - 09:00


HsDPA Traffic (Gb)

30

Monday

25 20 15

Sunday

10 5 0

7

8

9

10


11

12

13

14

15

16

17

18

19

20

21

22

23

Mon - 12:00 - 13:00

Mon - 20:00 - 21:00


HsDPA Traffic (Gb)

30

Monday

25 20 15

Sunday

10 5 0

7

8

9

10


11

12

13

14

15

16

17

18

19

20

21

22

23

Hotspots Detection #2 with single cell time-based analysis 5 days aggregation

College

Top 15 IMSIs distribution –College 8.6%

1

6.8%

2

5.8%

3

5.7%

4

5.4%

5


3.9%

3.7%

3.6%

2.8%

2.8%

2.6%

2.6%

2.5%

2.1%

2.1%

6

7

8

9

10

11

12

13

14

15

HetNet Planning Area Profile (RSCP, DCR, radio quality etc)

Geolocation

Network Load simulations

Devices Profile (traffic, signalling, avg throughput etc)

Logos + Net Probes

Logos

MUSA

Applications profile (apps used and RAN impact)

Churn network mapping

Churn map

Twitter, Flickr, Foursquare hot spot

Social Media users

Select the inputs based on SoW and requirements Driven Activities

Special Event Support

Traffic forecast & Capacity expansion

3G

Small cells

Small cells

In Building 6 sectors 17/09/2014 © Nokia 2014 - Geo-location - NPO Solutions

17 Confidential

DAS

Service Optimization

3D Geolocation Performance monitoring in buildings


Performance Monitoring In Buildings Provide network performances including traffic distribution, RF Events Positioning for Outdoors and indoors, in-building and at different floors of buildings.


Identify inbuilding performance issues

Monitor a critical building section Call drops 100+ 60-100 40-60 10-40 0-10

4260m

Events

36+ 25-36 21-25 14-21 11-14 10-11 8-10 6-8 5-6

Performance Monitoring #1 - Macro view % Ec/Io < (-14db), buildings, 0-21m Calculate the % of events that were measured at Ec/Io<(-14dB) for each building Analyze per building height, start with 0-21m

Numbers in brackets represent # of buildings per each category


Performance Monitoring #1 - Ec/Io<-14db, buildings, drill down by elevation (Normal) Within 10-30% of Ec/Io<-14db on low floors

Ec/Io<-14db, buildings, 0-21m

(Normal) Within 30-50% of Ec/Io<-14db on medium floors

Ec/Io<-14db, buildings, 42-60m 21 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

(Normal) Within 30-50% of Ec/Io<-14db on low floors

Ec/Io<-14db, buildings, 21-42m (Normal) Within 30-50% of Ec/Io<-14db on medium floors


Performance Monitoring #1 - Ec/Io<-14db, buildings, drill down by elevation (Normal) Within 30-50% of Ec/Io < -14db on high floors


(Bad) Within 60-70% of Ec/Io < -14db on highest floors

Ec/Io<-14db, buildings, 120-141m 22 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

(Normal) Within 30-50% of Ec/Io < -14db on high floors


Performance Monitoring #1 - Path Profile Analysis

Clear Line of Sight

Clear Line of Sight 23 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

Clear Line of Sight

Some intrusion, no Line of Sight

Performance Monitoring #1 - Path Profile Analysis

Some intrusion but still have Line of Sight to the upper floors


Performance Monitoring #2 - Call Performance by Indoor & Outdoor Traffic

Drops in buildings (total)

Drops on Buildings and Roads (by Ec)

Drops on Az Towers 4060m Height

Voice MoU Buildings Begin Rd. 25 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

3D Geolocation MV RF optimization / improving overall network quality KPI


RF Optimization/Improving Overall Network Quality KPI Identify indoor and outdoor installation problems with geolocation ‘audit’

Reduce interference through automatic optimization


Optimization Strategy Definition

NetOptim – Changes Improved

Spoiled

4523

646

Increased

Decreased

49568

26530

Case Study - Optimization Project results using Geo-location Customer’s challenge • Reduce drop calls without adding sites • Project Size: 101 Cells, 2100MHz • AOI 7 x 7Km • Guard Area 15 x 15Km • 45 changes in 25 sites (power, tilt)

Results


Decision Support to Implementation

RSCP – Before & After. Great improvement in problematic areas

RSCP Before

RSCP Before 29 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

RSCP After

RSCP After

EcIo – Before & After. Ec/Io levels above -10 dB were improved by 20%

EcIo Before

EcIo After

EcIo Before

EcIo After


Voice & HS Traffic - Before & After Based on OSS counters, Erlang was increased by 6%.

Voice Traffic Before

Voice Traffic After

Based on OSS counters, HS Payload was increased by 15% HS Traffic Before 31 17/09/2014 © Nokia 2014 - Geo-location - NPO Confidential

HS Traffic After

References

France

Germany

USA ME

Mexico

Hong Kong Singapore

Australia


LTE & MV Roadmap Geo-location engine Release 1.0: Current

Release 2.0: P11 2014

Release 3.0: P4 2015

Release 4.0: Q2 2015

GSM & UMTS

LTE

LTE

LTE

Planning and Optimization

Planning



Nokia, E//, Huawei and any other vendor (soft IP probes)

Nokia

Nokia

Nokia, E//, Huawei and other vendors



Small Cell Planning using Geolocation in Australia

Back

Customer’s challenge • Replace existing GSM micro cell sites covering Bondi beach area in Sydney with 3G and LTE co-located small cells • Small cells expected to provide traffic offloading without causing interference to macro layer • Location of sites pre-determined, imposing further constraints on RF design

Nokia’s solution • Feasibility study of deploying UMTS 2100 shared carrier small cells and LTE 2600 dedicated carrier as a capacity layer • User distribution map analysis based on geo-location used to identify hotspots and validating antenna direction • Coverage and capacity analysis


Benefits

• Achieved good ROI from small cells deployment using geolocation to drive Hetnet planning and Antenna placement • Small cells successfully offload traffic from macro U21 layer during the special event  Carried up to 35% of HSDPA traffic  Carried up to 62% of HSUPA traffic  Carried up to 13% of voice traffic

Radio Optimization using 3D Geolocation

Back

Customer’s challenge • Reduce drop call rate without adding sites in U2100 layer • RF shaping via OSS: change only CPICH power and electrical tilts • Reduce the cost for Drive-testing

Nokia’s solution • Monitoring maps on both roads and in-buildings based on 3D geolocation to analyse interference • Conduct antenna system configuration optimization and neighbor optimization • Demonstrate the use of Geolocation to replace Drivetesting


Benefits

• Improved efficiency by up to 80% reduction in drive testing • Significant KPI improvements in a matured network  Dropped Call Rates improved by 13.2%  Ec/Io levels above -10 dB were improved by 20%  Voice Traffic (Erlang) increased by 6%  HS Payload increased by 15%

Radio Optimization using 3D Geolocation in Singapore Customer’s challenge • Signal quality for both indoor and outdoor in ultra-dense business district area • Further improvement in call drop performance for a network with already good KPI level

Nokia’s solution • Achieve improvement targets set by customer • Monitoring maps on both roads and in-buildings based on 3D geolocation to analyse interference • Conduct antenna system configuration optimization and neighbor optimization


Benefits • • • •

Ec/Io levels above -10 dB were improved by 30% Dropped Call Rates (Outdoor) improved by 17% Dropped Call Rates (Indoor) improved by 10% Overall Call Drop improvement by 15%

Back

Monitoring using Geolocation in Europe Customer’s challenge • Estimate the need for small cells for 2014/15/16 taking into account traffic growth and spectrum strategy • Require location of traffic hot spots and a list of potential site candidates to support expansion plan

Nokia’s solution • Develop 3 traffic scenarios and identify the critical sectors in LTE and UMTS • Use of 3D geolocation solution to identify traffic hot spots and critical QoE areas


Benefits

• Detailed Capacity dimensioning plan within the budget definition • List of hot spots identified requiring small cell layer for offloading

07_Nokia 3D Geo-Location

Recommend Documents