FP7 ICT-SOCRATES

Handover parameter
optimization in LTE self-
organizing networks

72nd Vehicular Technology Conference

69 September 2010
Ottawa, Canada

T. Jansen, I. Balan, J. Turk

I. Moerman, T. Krner
 Outline

1. Introduction

2. Simulation environment and metrics

3. Initial performance studies

4. Handover optimisation SON algorithm

5. Simulation results

6. Conclusion

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

2/17
 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

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

3/17
 MATLAB LTE system-level simulator

Input data
Realistic SOCRATES scenario Start

1.5 km * 1.5 km area

Up to 78 cells
Read scenario
Microscopic traffic simulator data

Mobile users (cars) with different

speed (up to 50 km/h)
No End of Yes
Ray-Tracer Simulation? End
Pathloss information to best 30 cells
User position updates every 100 ms

Update Save results

RSRP/SINR Next step
Update RSRP/SINR
3dB shadow fading map
HO procedure HO algorithm
Handover procedure / algorithm

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

4/17
 Simulation metrics

Control
Control parameters Values
parameter
Hysteresis (0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5,
Hysteresis 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10 )
in [dB]
Time-to-Trigger (0 0.04 0.064 0.08 0.1 0.128 0.16
Time-to-Trigger 0.256 0.32 0.48 0.512 0.64 1.024
1.280 2.56 5.12) in [s]
Assessment metrics

Handover failure ratio Call dropping ratio

N HO _ fail N HO _ dropped
HPI HOF HPI DC
N HO _ fail N HO _ succ N HO _ accepted
Ping-Pong handover ratio

N HO _ pp
HPI HPP
N HO _ pp N HO _ npp N HO _ fail
WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

5/17
 Simulation metrics

System metrics

RSRP (Reference Signal Received Power)

cell transmit power Pc

pathloss Lue to the UE
shadow fading L fad with a standard deviation of 3dB

RSRPc,ue Pc Lue L fad

SINR (Signal to Interference Noise Ratio)

interfering cells N
N RSRPn ,ue
10
SINRc ,ue RSRPc ,ue 10 log10 10
n 1

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

6/17
 Initial performance studies

Objective Simulation parameter Value

Analyse the system behaviour Simulation time 200 [s]

and sensitivity Simulation step time 0.01 [s]

Find handover algorithm Simulation area (mobile users) 1.5 km * 1.5 km

approach Number of users 30

eNodeB transmit power 46 [dBm]

Simulation assumptions
Number of considered cells in the scenario 76
All resources are used in all
Measured cells (N) 21
cells (maximum interference)
Considered interfering cells for SINR
20
calculations
Simulation approach
Critical ping-pong handover time (T_crit) 5 [s]
Perform system simulations for
Handover execution time 0.25 [s]
all hysteresis and time-to-
trigger value combination SINR averaging window 0.1 [s]

(handover operating point) Min. SINR threshold - 6.5 [dB]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

7/17
 Call dropping behaviour

Call drops

0.8
Call dropping ratio

0.6

0.4

0.2

0
5
2 10
1 8
0.5
6
0.25
4
0.1
2
Time-to-Trigger [s] Hysteresis [dB]
0 0

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

8/17
 Handover performance weighting function

HP = w1 HPIHOF + w2 HPIHPP + w3 HPIDC

wx is the weight of the individual HPI

HPIHOF is the handover failure performance indicator

HPIHPP is the ping-pong handover performance indicator

HPIDC is the dropped calls performance indicator

Weighting parameter Value

w1 1.0
w2 0.5
w3 2.0

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

9/17
 Handover performance
Handover Performance (weights = [1 0.5 2])

Normalised sum of weighted HO failure rate,

ping-pong HO rate and call dropping rate

0.5

0
10
5
2 8
1
6
0.5
0.25 4
0.1
2
Time-to-Trigger [s] Hysteresis [dB]
0 0

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

10/17
 Simulation parameters for the performance analysis
Simulation parameter Value
Simulation time 1000 [s]
Simulation step time 0.01 [s]
Simulation area (mobile users) 1.5 km * 1.5 km
Number of users 50
eNodeB transmit power 46 [dBm]
Operating points (4, 0.48), (6, 0.32), (8, 0.1), (9, 0.08)
(Hysteresis, Time-to-Trigger) in [dB, s]

Number of considered cells in the scenario 78

Measured cells (N) 21

Considered interfering cells for SINR
20
calculations
Handover performance averaging window 60 [s]
Critical ping-pong handover time (T_crit) 5 [s]
Handover execution time 0.25 [s]
SINR averaging window 0.1 [s]
Min. SINR threshold - 6.5 [dB]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

11/17
 Performance of the non-optimised network

Handover Performance for the operating point (4, 0.48)

25
Handover failure
Ping-Pong handover
Call dropping
20

15
Ratio [%]

10

0
100 200 300 400 500 600 700 800 900 1000
Time [s]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

12/17
 Performance of the non-optimised network
Ping-Pong handover performance
Handover failure performance
25
4
Operating point (4, 0.48)
Operating point (4, 0.48)
Operating point (6, 0.32)
Operating point (6, 0.32)
3.5 Operating point (8, 0.1)
Operating point (8, 0.1) Operating point (9, 0.08)
20
Operating point (9, 0.08)
3

Ping-Pong handover ratio [%]

Handover failure ratio [%]

2.5 15

10
1.5

1
5

0.5

0 0
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
Time [s] Time [s]

Call dropping performance

6
Operating
Operating
point
point
(4,
(6,
0.48)
0.32)
Comparison of the network
Operating point (8, 0.1)
5
Operating point (9, 0.08) performance for four different
4 operating points
Call dropping ratio [%]

3
(4 dB Hys, 0.48 s TTT)
2
(6 dB Hys, 0.32 s TTT)
(8 dB Hys, 0.1 s TTT)
1

(9 dB Hys, 0.08 s TTT)

0
100 200 300 400 500 600 700 800 900 1000
Time [s]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

13/17
 Handover optimisation SON algorithm
HO SON algortihm
Optimisation criteria for HPIs
1) Handover Time-
Next cell Performance Hysteresis to- Optimisation
Indicator Trigger
2)
< 5 dB TTT
Update HPIs
Handover
5 dB 7 dB TTT & HYS
failure ratio
Yes
3)
No
> 7 dB HYS
HPIs <
threshold?
Ping-Pong < 2.5 dB TTT
4)
Increase good
6)
Increase bad
handover 2.5 dB 5.5 dB TTT & HYS
performance time performance time
ratio > 5.5 dB HYS
5) 7)
> 6 dB > 0.6 s TTT & HYS
Reset bad Reset good
performance time performance time
<= 6 dB > 0.6 s TTT
Call dropping > 7.5 dB <= 0.6 s TTT & HYS
Yes
8)
No
ratio 3.5 dB 6.5 <= 0.6 s HYS
Good
perform-
ance?
dB
9)
11)
< 3.5 dB <= 0.6 s TTT & HYS
Decrease HPI No
thresholds Bad
perform-
ance? Optimisation actions are added up
10)
Reset good
performance time 12)
Yes
Hys and TTT are only changed by one
Change handover
operating point step at a time
13) The new operating point has to belong to
the set of meaningful operating points
Reset bad
performance time

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

14/17
 Handover optimisation simulation results

Handover performance for the operating point (6, 0.32)

10
Handover failure
9 Ping-Pong handover
Call dropping
8

6
Ratio [%]

0
100 200 300 400 500 600 700 800 900 1000
Time [s]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

15/17
 Handover optimisation simulation results
Handover performance (Optimisation)
8
Handover failure
Ping-Pong handover
7
Call dropping

5
Ratio [%]

0
100 200 300 400 500 600 700 800 900 1000
Time [s]

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

16/17
 Conclusion

The system behaviour for 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 other scenario (done)
Problem: Fixed ratio of target thresholds between the HPIs
Enhance the handover optimisation algorithm (ongoing)
Introduce different user types (pedestrians, indoor, etc) (ongoing)

WWW.FP7-SOCRATES.EU Dipl.-Ing. Thomas Jansen, TU Braunschweig, Institut fr Nachrichtentechnik

17/17
FP7 ICT-SOCRATES

Thank you very

much for
your attention