Successive Interference Cancellation: A Back of the Envelope Perspective

1. Successive Interference Cancellation: A Back of the Envelope Perspective Souvik Sen, Naveen Santhapuri, Romit Roy Choudhury, Srihari Nelakuditi

2. Simple Case of Wireless Transmission AP T1 Decoding successful if: Signal Noise > Threshold SNR = =

3. What if parallel transmissions? AP T1 T2 Interferer Decoding successful only if: Signal Interference + Noise > Threshold SINR = =

4. Collision Collision AP T1 T2 Interferer Decoding fails when: Signal Interference + Noise < Threshold SINR = =

5. Successive Interference Cancellation AP T1 T2 Interferer 1. Decode strongest signal first

6. Successive Interference Cancellation AP T1 T2 Interferer 1. Decode strongest signal first 2. Model and subtract

7. Successive Interference Cancellation AP T1 T2 Interferer 1. Decode strongest signal first 2. Model and subtract 3. Normal Decode It is as if SIC can “uncollide” signals, resulting in two successful transmissions

8. Sblue Sgreen Sblue +Sgreen R*green = RSIC = log 1+ log 1+ Rblue = log 1 + SINR = SNR = noise noise Sblue +noise Capacity with SIC T2 Interferer AP T1 Blue bit rate remains same Green bit rate has to be far less Strong signal penalized, weak signal gets all the benefits

9. Gainsic = Sgreen Sblue Sblue +Sgreen Rgreen = log 1 + RSIC = log 1+ Rblue = log 1 + noise SNR = SNR = noise noise Channel Capacity w/o SIC T2 Interferer AP T1 RwoSIC = max( Rblue, Rgreen )

10. SIC Capacity Gain

11. SIC PHY Capacity Gain Max SIC capacity gain when equal signal strengths

12. Our interpretation was that ... maximizing SIC capacity will maximize throughput As MAC protocol designers ... We were tempted to schedule packet transmissions of similar signal strengths ...

13. T2 Interferer AP HOLE T1 SIC: A Packet Perspective • Weaker blue packet can be at a high rate • Stronger green packet has to be at low rate Rate Packet Transmission Time MAC Layer throughput can actually suffer

14. L L L Transmission Time T2 Interferer L Transmission Time AP + = , = max Rblue Rblue Rgreen T1 R*green Mathematically ... TimeSIC = TimewoSIC=

15. GainSIC = L L L T2 Interferer Transmission Time L Transmission Time AP + , max Rblue Rblue Rgreen T1 R*green Mathematically ... TimeSIC = = TimewoSIC= =

16. SIC Throughput Gain

17. SIC Throughput Gain Max throughput gain when signal strengths are 2:1

18. Capacity Vs. Throughput • We expected: • Maximizing SIC capacity will immediately maximize throughput • Reality: • Equal signal strengths maximize capacity • Disparate signal strengths (2:1) maximize throughput Capacity

19. Can’t we improve MAC layer throughput with SIC • by reducing size of the hole? • Certainly possible: • Power control • Scheduling • Multirate packetization • Packet packing

20. Can’t we improve MAC layer throughput with SIC • by reducing size of the hole? • Certainly possible: • Power control • Scheduling • Multirate packetization • Packet packing But at what cost?

21. We study SIC enabled throughput in two scenarios 1. Common receiver 2. Distinct receivers

22. We begin with 1. Common receiver

23. = 2 * (1) Power Control • Reduce power of blue Tx such that Rgreen SINR*green = Rblue Reduce

24. (2) Client Pairing T4 T1 T2 T3 T1, T2 T3, T4

25. (2) Client Pairing T4 T1 T2 T3 T1, T3 T2, T4

26. (3) MultiRate Packetization • Multirate Packetization • Send the strong packet at high rate after weak packet has finished R*green Rblue Rblue Rgreen

27. (4) Packet Packing • Packet Packing • Send multiple packets to fill up the hole • Hard because stronger signal modeling becomes difficult R*green Rblue

28. Monte Carlo Simulations SIC Packing Power Control Rate

29. Monte Carlo Simulations SIC Packing Power Control Rate Considerable Improvement with Adaptation

30. 2. Distinct receivers

31. 2. Distinct receivers R1 R2 T2 T1 • Main Concern: • Bit Rate of T1R1 is optimal • R2 has to decode T1’s signal at this bit rate • Despite the presence of T2’s signal

32. Gains available when several topological constraints hold: R1 R2 T2 T1 How often do these SIC permissible topologies occur?

33. Monte Carlo Simulations (AP Transmit Range) Gain with SIC in less than 10% of the cases

34. Does MAC Adaptation Help? Not many topologies support SIC … thus limited scope for protocols

35. Implication on Network Architectures?

36. R2 R1 T2 T1 • Enterprise WLANs: • Clients likely to associate with stronger AP • Such scenarios unlikely • Residential WLANs: • Neighbors AP may be stronger • Some SIC scenarios possible

37. Conclusion • Successive Interference Cancellation • A PHY layer capability to “uncollide” transmissions • Throughput gain not immediate from SIC • Permissible bit rates impact the length of packet transmission times • Creates under-utilization of the channel • Protocol adaptations possible to cope with problem • Some gains available for common receiver scenarios • However, limited gains for networks with distinct receivers

38. Take Away Message: SIC aware protocol design fraught with pitfalls … Consider doing a back-of-the-envelope calculation before plunging into system design

39. Thank You Duke SyNRG Research Group http://synrg.ee.duke.edu