Greedy algorithms for time frequency allocation in in a SDMA satellite communications system. Erwan CORBEL (Thales)

Transcription

1 Greedy algorithms for time frequency allocation in in a SDMA satellite communications system Kata KIATMANAROJ, Christian ARTIGUES, Laurent HOUSSIN (LAAS), Erwan CORBEL (Thales) 1

2 Contents Problem definition Greedy algorithms Experimental results Conclusions and perspectives 2

3 Problem definition 3

4 Problem definition To assign a limited number of frequencies to as many users as possible within the service area 4

5 Problem definition To assign a limited number of frequencies to as many users as possible within the service area Frequency is a limited resource! Frequency reuse > > co channel channel interference Intra system interference 5

6 Problem definition To assign a limited number of frequencies to as many users as possible within the service area Frequency is a limited resource! Frequency reuse > > co channel channel interference Intra system interference Graph coloring problem NP hard 6

7 Problem definition Interference constraints Binary interference Cumulative interference i j i j k 7

8 Problem definition Satellite beam & antenna gain SDMA: Spatial Division Multiple Access j i k 8

9 Problem definition Assignment Logical boxes (superframes) Demand = F x T No overlapping within the superframe Overlapping between superframes (simultaneous) may create interference o ij 1 9

10 Problem definition Superframe structure 10

11 Problem definition Frames and satellite beams 11

12 Problem definition Frame structure constraints 12

13 Problem definition Terminal types 50 dbw, 45 dbw Max. 24 Mbps, 10 Mbps Traffic types Guaranteed, Non guaranteed 13

14 Problem definition User priority level and handling 0 3 Weighted Round Robin Robin ordering 14

15 Problem definition Symbol rate Modulation Coding scheme (RsModCod) 16 ModCod 4 symbol rates (Rs) corr. to 5, 10, 15 and 20 MHz Support bitrate (Mbps) Different acceptable interference thresholds (alpha) 15

16 Problem definition Beam positioning methods Fixed beam SDMA beams 16

17 Problem definition Uplink power control After the resource assignment PCMargin Overall interference reduction 17

18 Greedy algorithms 18

19 Greedy algorithms Fast Flexible No look back principle Extensive hierarchical search MI (Minimum Interference) MB (Minimum i Bandwidth) No performance guarantee 19

20 ILP vs. Greedy (single carrier case) Performance comparison: ILP vs. Greedy Numb ber of accepted users Greedy ILP (60s) ILP (180s) Number of users 20

21 Greedy algorithms: MI NbS (superframe) m n (bin configurations) y1 y2 (low high frequencies) x1 x2 (leftmost rightmost time bin) Interference calculation repeats * Use control parameters to limit the search space 21

22 Greedy algorithms: MI Minimum Interference (MI) New superframe when the old one is utilized. MI Superframe 1 Superframe 2 22

23 Greedy algorithms Minimum Interference (MI) 23

24 Greedy algorithms: MB m n y1 y2 x1 x2 NbS Interference calculation repeats 24

25 Greedy algorithms Minimum Bandwidth (MB New superframe before increasing bandwidth 25

26 Experimental results 26

27 Computational experiments Test instances 27

28 Experimental results Assignment time (seconds) 28

29 Experimental results Number of rejected users 29

30 Experimental results Frequency utilization (MHz) Note: system maximum bandwidth 300 MHz 30

31 Conclusions and further study Highly complex problem and fast calculation time requirement ILP impractical MI: least interference MB: least bandwidth Lower bounds on the number of rejected users Local search heuristics 31

32 Thank you 32

33 Experimental results Total interference gap 33

34 Experimental results Number of optima for ILPs 34