FP7 ICT-SOCRATES Handover parameter optimization in LTE selforganizing networks TD (1)168 COST 1, 1 th MCM Athens, Greece February 3 rd 5 th TUBS, Braunschweig, Germany IBBT, Ghent, Belgium VOD, Newbury, England
Outline 1. Introduction. Simulation scenario and LTE system-level simulator 3. Simulation metrics 4. Controllability and Observability studies 5. Performance of the non-optimised network 6. Handover optimisation SON algorithm 7. Simulation results 8. Conclusion /
Introduction Problem Handover parameter optimisation is done manually high OPEX long optimisation intervals based on error reports Non-optimal handover performance handover failures ping-pong handovers call dropping Handover parameter optimisation objective automate the optimisation adapt the handover parameters on a short-term scale optimise the handover performance Approach analyse the system behaviour develop handover optimisation algorithm 3/
Realistic SOCRATES Scenario Simulations LTE Simulator Assembling Scenario Data Network Information Decorated User Snapshots Processing Data Correlated User Snapshots Network Environment User locations Generating Source Data Network data OpenSteetMap Braunschweig Scenario Traffic Distribution 4/
Realistic SOCRATES Scenario Computing the landuse information from openstreetmap.org Landuse classes: Road, Building, Water, Street and Railway 5/
MATLAB LTE system-level simulator Input data Realistic SOCRATES scenario Start Read input data Power mask Build Network Build Users Soft frequency reuse Call generation End of Simulation? No Set Power Mask All users connected Yes End Call Generation Update RSRP/SINR Shadow fading maps Next step Update RSRP/SINR Handover procedure/algorithm HO algorithm HO procedure 6/
Simulation metrics Control parameters Hysteresis Time-to-Trigger Assessment metrics Control parameter Hysteresis Time-to-Trigger Values (,.5, 1, 1.5,,.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 1 ) in [db] (.4.64.8.1.18.16.56.3.48.51.64 1.4 1.8.56 5.1) in [s] Handover failure ratio HPI HOF N HO_ N fail HO_ fail N HO_ succ Call dropping ratio HPI DC N N HO_ dropped HO _ accepted Ping-Pong handover ratio HPI HPP N HO_ pp N N HO_ pp HO_ npp N HO_ fail 7/
Simulation metrics System metrics RSRP (Reference Signal Received Power) cell transmit power L ue pathloss to the UE L fad P c shadow fading with a standard deviation of 3dB RSRP, c ue P c L ue L fad SINR (Signal to Interference Noise Ratio) SINR interfering cells N N RSRPn, ue 1 c, ue RSRPc, ue 1 log1 1 n 1 8/
Controllability and Observability studies Objective Analyse the system behaviour and sensitivity Find handover algorithm approach Simulation parameter Value Simulation time [s] Simulation step time.1 [s] Simulation area (mobile users) 1.5 km * 1.5 km Number of users 3 Simulation assumptions All resources are used in all cells (maximum interference) enodeb transmit power 46 [dbm] Number of considered cells in the scenario 76 Measured cells (N) 1 Simulation approach Perform system simulations for all hysteresis and time-totrigger value combination (handover operating point) Considered interfering cells for SINR calculations Critical ping-pong handover time (T_crit) Handover execution time SINR averaging window Min. SINR threshold 5 [s].5 [s].1 [s] - 6.5 [db] 9/
C & O: Handover failures Handover failure ratio Handover Failures.4. 1 8 5 6 1.5.5 Time-to-Trigger [s].1 4 Hysteresis [db] 1/
C & O: Ping-Pong handovers Ping-Pong handover ratio Ping-Pong Handovers.8.6.4. 5 1 1.5.5.1 4 6 8 Time-to-Trigger [s] Hysteresis [db] 11/
C & O: Call dropping Call dropping ratio Call drops.8.6.4. 5 1.5.5 Time-to-Trigger [s].1 8 6 4 Hysteresis [db] 1 1/
Handover performance weighting function HP = w 1 HPI HOF + w HPI HPP + w 3 HPI DC w x is the weight of the individual HPI HPI HOF is the handover failure performance indicator HPI HPP is the ping-pong handover performance indicator HPI DC is the dropped calls performance indicator Weighting parameter Value w 1.5,.6,,. w.5,.6,,. w 3.5,.6,,. 496 valid weighting parameter combinations have been considered If (HP<.5) => meaningful handover parameter operating point 13/
Handover performance Normalised sum of weighted HO failure rate, ping-pong HO rate and call dropping rate Handover Performance (weights = [1.5 ]) 1.5 5 1 1.5.5.1 4 6 8 Time-to-Trigger [s] Hysteresis [db] 14/
meaningful handover operating points Operating Points (Threshold: 5%) 5 1.5.5 Time-to-Trigger [s].1 1 3 4 8 7 6 5 Hysteresis [db] 9 1 15/
Simulation parameters for the performance analysis Simulation parameter Simulation time Simulation step time Simulation area (mobile users) Number of users 5 enodeb transmit power Operating points (Hysteresis, Time-to-Trigger) Number of considered cells in the scenario 78 Measured cells (N) 1 Considered interfering cells for SINR calculations Handover performance averaging window Critical ping-pong handover time (T_crit) Handover execution time SINR averaging window Min. SINR threshold Value 1 [s].1 [s] 1.5 km * 1.5 km 46 [dbm] (4,.48), (6,.3), (8,.1), (9,.8) in [db, s] 6 [s] 5 [s].5 [s].1 [s] - 6.5 [db] 16/
Performance of the non-optimised network Ratio [%] 5 Handover Performance for the operating point (4,.48) Handover failure Ping-Pong handover Call dropping 15 1 5 1 3 4 5 6 7 8 9 1 Time [s] 17/
Call dropping ratio [%] Handover failure ratio [%] Ping-Pong handover ratio [%] 3.5 Performance of the non-optimised network 4 3 Operating point (4,.48) Operating point (6,.3) Operating point (8,.1) Operating point (9,.8) Handover failure performance 5 Ping-Pong handover performance Operating point (4,.48) Operating point (6,.3) Operating point (8,.1) Operating point (9,.8).5 15 1.5 1 1 5.5 1 3 4 5 6 7 8 9 1 Time [s] 1 3 4 5 6 7 8 9 1 Time [s] 6 5 4 3 1 Operating point (4,.48) Operating point (6,.3) Operating point (8,.1) Operating point (9,.8) Call dropping performance 1 3 4 5 6 7 8 9 1 Time [s] Comparison of the network performance for four different operating points (4 db Hys,.48 s TTT) (6 db Hys,.3 s TTT) (8 db Hys,.1 s TTT) (9 db Hys,.8 s TTT) 18/
Handover optimisation SON algorithm 4) 5) 9) 1) 1) ) Increase good performance time Reset bad performance time Decrease HPI thresholds Reset good performance time HO SON algortihm Next cell Update HPIs 3) Yes No HPIs < threshold? 6) 7) 8) Yes Good performance? No 1) 13) Increase bad performance time Reset good performance time 11) Yes Change handover operating point No Reset bad performance time Handover Performance Indicator Handover failure ratio Ping-Pong handover ratio Call dropping ratio Optimisation criteria for HPIs Bad performance? Hysteresis Time- to- Trigger Optimisation < 5 db TTT 5 db 7 db TTT & HYS > 7 db HYS <.5 db TTT.5 db 5.5 db TTT & HYS > 5.5 db HYS > 6 db >.6 s TTT & HYS <= 6 db >.6 s TTT > 7.5 db <=.6 s TTT & HYS 3.5 db 6.5 <=.6 s HYS db < 3.5 db <=.6 s TTT & HYS Optimisation actions are added up Hys and TTT are only changed by one step at a time The new operating point has to belong to the set of meaningful operating points 19/
Handover optimisation simulation results Ratio [%] 1 9 8 Handover performance for the operating point (6,.3) Handover failure Ping-Pong handover Call dropping 7 6 5 4 3 1 1 3 4 5 6 7 8 9 1 Time [s] /
Handover optimisation simulation results Ratio [%] 8 7 Handover performance (Optimisation) Handover failure Ping-Pong handover Call dropping 6 5 4 3 1 1 3 4 5 6 7 8 9 1 Time [s] 1/
Conclusion The system behaviour to different handover operating points has been analysed Handover performance can be optimised using the proposed algorithm Handover operating points are chosen for every cell individually The overall network performance is increased and the handover failure ratio and ping-pong ratio drop to zero in the shown case Next steps Run the algorithm in a larger scenario Improve the SINR calculation (scheduling) Introduce background traffic (implication on system throughput) User specific handover parameters /
FP7 ICT-SOCRATES Thank you very much for your attention
Handover procedure I HO procedure 1) Next active user No ) HO command send? Yes 3) Find the best server 8) Send Handover command Yes 4) Connected to the best server? No 5) Save best server as HO candidate 6) If new cell is best server set back HO crit. time 7) Increase handover criteria time Yes HO criteria time > TTT? No 4/
Handover procedure II 1) Next active user No 17) HO failure occured? No 1) Reconnect to Source enode B 1) Increase HO duration time No 11) HO duration > HO execution? Yes 18) Save Handover failure 19) Hand back successful? Yes ) Save call drop during Handover Yes 16) Save Ping-Pong Handover Yes 1) Handover complete 13) Save successful Handover 14) Update UE History 15) Ping-Pong HO detected? No The handover procedure is executed in every simulation time step Handover procedure is independent of the handover algorithm 5/