1. Naveen Santhapuri, Srihari Nelakuditi and Romit Roy Choudhury University of South Carolina Duke University WCNC 2008

2. Outline Introduction Related work CAST Simulation Conclusions

3. Introduction The optimal network capacity for wireless networks Related to spatial reuse Efficient spatial reuse is inhibited by Interference External noise Other physical factors Limitations of MAC protocols

4. Introduction The 802.11 protocol with its virtual carrier sensing has role reversals which Reduce the hidden node problem Introduce the exposed node problem

5. Introduction If there was no ACK Two nodes can transmit DATA simultaneously Exposed sender problem can be solved partially

6. Piggybacked ACK mechanism “Piggybacked-Ack-aided Concurrent Transmissions” N. Santhapuri, J. Wang, Z. Zhong, and S. Nelakuditi ICNP Poster Session, 2005 ABC DATARTS CTS for A ACK for C

7. Introduction -- collision Multiple packets arriving at a receiver are generally considered to cause packet loss due to the collision at the receiver ABC

8. Introduction A sufficiently stronger frame can still be successfully received by the receiver in spite of a collision This phenomenon is called physical layer capture (PLC)

9. Related work “Sniffing out the correct physical layer capture model in 802.11b” A. Kochut, A. Vasan, A. U. Shankar and A. Agrawala ICNP, Oct. 2004 R1S1S2R2

10. ICNP-2004 S1 R1 S3 R1S1S2R2S3 S1 R1 S2 PreambleDATA XX 100m 50m

11. Related work “An experimental study on the capture effect in 802.11a networks” J. Lee, W. Kim, S.-J. Lee, D. Jo, J. Ryu, T. Kwon and Y. Choi WinTECH, Sept. 2007

12. WinTECH-2007

13. A signal is significantly more vulnerable to interference if it starts after the interfering frame than had it started before the interfering frame. S1 R1 S3 S1 R1 S2 interest interfering X interest interfering X PreambleDATA

14. Proposed approach Assumption Assume two hop signal strength information S1’s signal strength at R1 will be known by S2 R1S1S2 100m 50m

15. Proposed approach Capture-Aware Staggering of Transmissions (CAST) S1 ->R1 RTS DATA R1S1S2R2S3 100m 50m 802.11 CAST S1 ->R1 S2 ->R2 ACK CTS PWAIT

16. CAST RTS: S1 -> R1 No other frame > SF hold S1 ->R1 RTS DATA 802.11 CAST S1 ->R1 S2 ->R2 ACK CTS PWAIT The SINR value at R1 for a signal from S1 in the presence of interference from S2 The SINR value at R1 for a signal from S1 in the presence of interference from S2

17. CAST CTS: R1 -> S1 RTS: S2 -> R2 after one physical preamble CTS starts Medium is free > SF hold RTS starts > SLC hold S1 ->R1 RTS DATA 802.11 CAST S1 ->R1 S2 ->R2 ACK CTS PWAIT

18. CAST CTS: R2 -> S2 DATA: S1 -> R1 after 2 physical preamble times CTS starts Medium is free > SF hold DATA starts > SLC hold S1 ->R1 RTS DATA 802.11 CAST S1 ->R1 S2 ->R2 ACK CTS PWAIT

19. CAST DATA: S1 -> R1 DATA: S2 -> R2 S1 -> R1 Medium is free > SF hold S2 -> R2 > SLC hold S1 ->R1 RTS DATA 802.11 CAST S1 ->R1 S2 ->R2 ACK CTS PWAIT

20. CAST In 802.11, for two concurrent transmissions 8 signal strength value must be above SLC threshold In CAST, for two concurrent transmissions Only 5 signal strength value must greater than SLC

22. Optimize the protocol RTS DATA CAST Primary Secondary CTS PWAIT CAST Primary Secondary

24. Simulation Qualnet simulator Traffic flow: random Traffic rate: CBR Packet size: 512 Bytes

25. Grids in a fixed sized area Space: 1000m*1000m Each grid has a different grid unit 75m ~ 175m

26. Random Topologies Space: 1000m*1000m Nodes: 100 Randomly generated 40 1-hop flows

27. Conclusions Spatial reuse in wireless networks is limited by the SINR threshold requirements This problem is amplified because of role reversals in wireless networks Physical layer capture can improve the spatial reuse by staggering the transmissions Simulation results show that the number of concurrent transmissions can be improved significantly

28. Thank you