How to Minimize the Impact of Cell Breathing on UMTS Networks IEEE Workshop on Applications and Services in Wireless Networks 2002 July 3 rd - 5 th, 2002 Yannick DUPUCH Alcatel - Mobile Networks Division GSM/UMTS Project Manager
Agenda Facing the Multimedia traffic increase Cell breathing and cell range evolution Strategic Cell Radius Upgrade your Network Advanced Radio Features Elaborating the deployment strategy How to minimize the impact of cell breathing effect on UMTS network / July 4th / 2
Cell Breathing Multiservice Traffic characteristic Various data rates (from Voice 12.2kbps up to 384Kbps) Various QoS and GoS (blocking, delay, throughput, BLER) Various connection types (Real Time and Non Real Time) Various traffic asymmetry and activity behaviour. Multiservice Multiservice Different Different Data Data rates rates Different Sensitivities Different Sensitivities 3G UE Voice (CS/2%) File transfer (PS/5s) Web browsing (PS/3s) Video conference (CS/2%) Shopping on line (PS/1s) Interactive games (CS/1%) Air interface Node B Different Different Footprints Footprints Different footprints for different services How to minimize the impact of cell breathing effect on UMTS network / July 4th / 3 High Data Rate (eg 384kb/s) Voice and Low Data Rate (eg up to 64kb/s) Medium Data Rate (eg 144kb/s)
Cell Breathing Multiservice Environment Seamless Coverage 100 subs/km 2 - Year 1 Coverage for each service is traffic dependent Cell shrinkage Service Coverage: Blue: Voice Only Green: Voice and CS64 Red: Voice, CS 64 and PS384 300 subs/km 2 - Year 3 How to minimize the impact of cell breathing effect on UMTS network / July 4th / 4 1000 subs/km 2 - Year 5
From traffic forecast to coverage prediction Iterative Link Budget Dimensioning Multiservice Traffic inputs W-CDMA Radio Parameters Cell Range Range assumption TRAFFIC ANALYSIS TRAFFIC ANALYSIS UL traffic DL traffic Interference Iterative process until convergence LINK BUDGET ANALYSIS LINK BUDGET ANALYSIS UL analysis DL analysis Final UL & DL Cell Range UL & DL Cell range > Use of Multiservice traffic modelling and iterative interference calculation in both uplink and downlink to get the evolution of the cell range How to minimize the impact of cell breathing effect on UMTS network / July 4th / 5
Strategic cell radius Fixing a time variant parameter > Strategic cell range will define the deployed inter-site distance per environment Site Site Site Site Site Site Site 1.5xR Cell range (km) 0.65 0.6 0.55 Dense Urban / Cell range = f(phases) Example of strategic cell radius Traffic increases PHO PH1 PH2 PH3 Phases How to minimize the impact of cell breathing effect on UMTS network / July 4th / 6 Voice with 1 carrier Seamless coverage Holes in coverage
Advanced Features Avoiding Cell Breathing problems by timely upgrade > Anticipate holes through coverage prediction > Avoid holes by timely upgrade In case of Downlink limitation TX Diversity High power amplifier In case of Uplink limitation Tower Mounted Amplifier (TMA) 4 Way RX Diversity Multi User Detection, MUD For either Uplink or Downlink limitation: Adding carriers Sectorisation Adding sites (densification) Smart antennas How to minimize the impact of cell breathing effect on UMTS network / July 4th / 7
Tower Mounted Amplifier Uplink Coverage Gain 18 Antenna Duplexer TMA Tx Rx Duplexer Feeder Tx / Rx Total Interference I (db) 16 14 12 10 8 6 4 Link Budget Curve with TMA Link Budget Curve w/o TMA I(R) for High_Traffic I(R) for Low_Traffic Typical reduction of the required number of sites: ~40% for low traffic scenario ~30% for high traffic scenario 2 Node B 0 0 0.2 0.4 0.6 0.8 1 How to minimize the impact of cell breathing effect on UMTS network / July 4th / 8 Cell Range R (km)
4-RX diversity on Uplink Uplink Coverage Gain 10 9 8 Polarisation Diversity Space Diversity Polarisation Diversity Total Interference I (db) 7 6 5 4 3 2 1 0 0 0,2 0,4 0,6 Cell Range R (km) Xpol antenna 1 Typical reduction of the required number of sites: Distance d Rxdiv1 Rxdiv2 Rxdiv3 Rxdiv4 ~30% for low traffic scenario ~40% for high traffic scenario Xpol antenna 2 Link Budget Curve w/o 4RxDiv Link Budget Curve with 4RxDiv I(R) for High_Traffic w/o 4RxDiv I(R) for High_Traffic with 4RxDiv I(R) for Low_Traffic w/o 4RxDiv I(R) for Low_Traffic with 4RxDiv How to minimize the impact of cell breathing effect on UMTS network / July 4th / 9
High Power Amplifier Impact of Node B power on DL capacity 45 Transmit power Transmit (Watt) power (Watt) per carrier 40 35 30 25 20 15 10 5 RURAL 7 km RURAL 5 km SUBURBAN 1,3 km URBAN 0,5 km URBAN DENSE 0,35 km Rural Capacity gain 0 100 200 300 400 500 600 700 800 900 Throughput NRT 128 (kbps) Sector Capacity Urban > Negligible impact in urban areas > Visible gain only in low density rural areas How to minimize the impact of cell breathing effect on UMTS network / July 4th / 10
High Power Amplifier DL Gain in multicarrier configurations - Urban 50% 20W vs. 35W at antenna connector: NRT 128 in Urban 40% From 6.7W per carrier to 11.7W per carrier (3 carriers) Capacity gain (%) 30% 20% From 10W per carrier to 17.5W per carrier (2 carriers) From 20W per carrier to 35W per carrier (1 carrier) small gain for 3 carriers Uplink Coverage limited cell ranges 10% negligible gain for 1 carrier 0% 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Cell range (km) How to minimize the impact of cell breathing effect on UMTS network / July 4th / 11
TX Diversity on Downlink Performance Gain > Performance gain of TX diversity feature consists of three aspects: Doubling the TX power by adding a power amplifier TX Bus TX1 TX1div Power Amplifier PA PA TEU 20 W 20 W Antenna Antenna 1 Antenna 2 TEU Reducing the required transmit power for each downlink channel (transmit power raise due to fast fading is reduced) Improving the required received E b /N 0 (slight reduction for open loop TxDiv, higher for closed loop TxDiv) How to minimize the impact of cell breathing effect on UMTS network / July 4th / 12 Target Rx Eb/N0 (db) 9 8 7 Speech 8 kbps, 1 rx antenna, downlink, pedestrian A Without Tx diversity STTD 0.8 db 6 3 6 10 25 50 120 Speed (km/h)
TX Diversity on Downlink Gain on Downlink Capacity - Example Capacity gain by doubling the max. downlink transmit power (%) > Gain highly dependent on diversity technique (open or closed loop) and multipath profile (Pedestrian>>Vehicular) 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% Example: STTD - Space-Time transmit diversity (open loop) Vehicular A, NRT 128, Urban, From(20W,1Carrier) To (40W,1Carrier) Uplink Coverage limited cell ranges 0% 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Pure Diversity gain in capacity: ~8% Gain through 2nd PA: dependent on cell range (curve) Example for typical cell range (0.55km): 8%+3%=11% total gain Cell range (km) How to minimize the impact of cell breathing effect on UMTS network / July 4th / 13
Adding a carrier Uplink coverage 16 14 > Uplink Coverage: Total Interference I (db) 12 10 8 6 4 2 link budget curve I(Traffic),1 carrier I(Traffic), 2 Carriers Link Budget curve stays the same, traffic curve depends on # of carriers 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Cell Range R (km) UL Coverage gain Low traffic High traffic Carrier config Dense Urban Rural Dense Urban Rural 1C -> 2C 5% 10% 20% 20% How to minimize the impact of cell breathing effect on UMTS network / July 4th / 14 2C -> 3C 3% 5% 15% 15%
Adding a carrier Downlink capacity 100% Significant increase of downlink capacity Exact gain depends on the hardware configuration (Nb of PA per sector, # of carriers, etc ) and cell range Carrier Adding a carrier implies: reducing power per carrier TX C1 (20W 2x10W) PA C2 Power Amplifier Antenna Antenna 1 10 W per carrier Capacity gain (%) 75% 50% 25% 0% From 1 carrier (20W p.c.) to 2 carriers (10W p.c.) From 2 carriers (10W p.c.) to 3 carriers (6W p.c.) Uplink Coverage limited cell ranges How to minimize 0 the impact of 0.2 cell breathing effect 0.4 on UMTS network / 0.6 July 4th / 15 0.8 Cell range (km) Adding a carrier is not only a downlink feature, but will bring for fixed intersite distance uplink capacity gain as well Most powerful enhancement feature
Deployment strategy Incumbent Operator GSM Cell range=strategic cell radius Start: Phase 0 Coverage and Capacity analysis Traffic forecast Holes or capacity problems? NO Apply UL feature (TMA, 4RXDiv, MUD) (if not yet used) YES Which limitation? UL DL both UL&DL Apply DL feature (TxDiv, HPA) (if not yet used) Add a carrier (if still available) Look at next phase Maximize the re-use of existing GSM sites How to minimize the impact of cell breathing effect on UMTS network / July 4th / 16
Deployment strategy Greenfield Operator Evaluate Cell Ranges (for all traffic phases and scenarios) Identify suitable strategic ranges Apply possible deployment scenarios (service deployment, advanced features upgrade) There is no unique solution for strategic cell range and deployment strategy, the final choice is an (individual) strategic decision! Build business cases Select the solution that best fits your business requirements How to minimize the impact of cell breathing effect on UMTS network / July 4th / 17
Conclusion > Analytical dimensioning techniques can be used for predicting and avoiding potential problems due to cell breathing ( upgrade strategy) > The network evolution can be considered already at dimensioning stage > The upgrade plan is not finished at dimensioning stage: continuous adaptation is required through Monitoring traffic Regularly updating the traffic forecast Adjusting the upgrade strategy How to minimize the impact of cell breathing effect on UMTS network / July 4th / 18
Thank you for your attention Yannick DUPUCH UMTS Project Manager Mobile Networks Division yannick.dupuch@alcatel.fr