Understanding Congestion Control in Multi-hop Wireless Mesh Networks

1. Understanding Congestion Control in Multi-hop Wireless Mesh Networks SumitRangwala ApoorvaJindal, Ki-Young Jang, KonstantinosPsounis, and RameshGovindan

2. Mesh Networks • Static multi-hop mesh networks have been proposed an alternative to wired connectivity • User’s satisfaction hinges on transport performance • TCP’s performance on 802.11 mesh networks is known to be poor • Starvation Is poor transport performance inherent to multi-hop mesh networks? Can a correctly designed transport help make mesh networks a viable alternative?

3. TCP’s Performance 1 3 2 4 5 6 8 • TCP only signals flows traversing the congested link • Link centric view of congestion • Fails to account for neighborhood congestion 7 9 What mechanisms can help us achieve near-optimal rates? Optimal (Max Min) TCP

4. Approach Neighborhood-centric Transport AIMD Based Design Explicit Rate Notification WCP WCPCap

5. Neighborhood of a Link Link →sender receiver pair • Neighborhood of a link • All incoming and outgoing links of • Sender • Receiver • One hop neighbors of the sender • One hop neighbors of the receiver 9 1 7 2 5 6 8 4 10 3 Prohibits channel capture at the sender or causes collision at the receiver Ensuing ACK prohibits channel capture at the sender or causes collision at the receiver Prohibits channel capture Neighbors (overhearing)

6. WCP: AIMD Based Design Multiplicative Decrease When a link is congested, signal all flows traversing the neighborhood of a linkto reduce their rate by half, i.e., rf = rf / 2 React to congestion after RTTneighborhood Key Insight: Congestion is signaled to all flows traversing neighborhood of a congested link

7. WCP Additive Increase During no congestion increase a flow’s rate as rf = rf + α Every RTTneighborhood RTTneighborhood : Largest flow RTT within the neighborhood Key Insight: Rate adaptation is clocked at the largest flow RTT in a neighborhood

8. Simulations: Stack Topology • Simulation setup • Qualnet 3.9.5 • 802.11b MAC with default parameters • TCP SACK • Auto rate adaptation is off • WCP achieves near optimal performance • Through congestion sharing in the neighborhood 1 3 2 4 5 6 8 7 9

9. Approach Neighborhood-centric Transport AIMD Based Design Explicit Rate Notification WCP WCPCap

10. WCPCap: Explicit Rate Feedback • Estimate residual capacity in a neighborhood • Need to know the achievable rate region for 802.11-scheduledmesh networks • Using only local information • Challenge: Is a given set of rates achievable in a neighborhood?

11. Calculating Achievable Rates Check feasibility, i.e., for each link, Packet arrival rate × E[service time of a packet] ≤ U, 0 ≤ U ≤ 1 Compute expected packet service time for a link from collision and idle probability of the link Decompose the neighborhood topology of a link into canonical two-link topologies Find collision and idle time probability of the link in every two-link topology Combine, incorporating local link dependencies, individual probabilities to find net collision and idle probabilities for the link Combine, incorporatinglink dependencies, individual probabilities to find net collision and idle probabilities of the link Using only local information Requires global information Jindal et. al., “The Achievable Rate Region of 802.11 Scheduled Multi-hop Networks”.

12. WCPCap: Explicit Rate Feedback • Every epoch • Find, by binary search, the largest increment or smallest decrement, δ, such that the new rates are achievable yet fair • Increase/decrease rate of each flow by δ U=1 (100% utilization) would yield large delays, we target U=0.7

13. Simulations: Stack Topology 1 3 2 4 5 6 8 • Simulation setup • Qualnet 3.9.5 • 802.11b MAC with default parameters • TCP SACK • Auto rate adaptation is off • WCPCap slightly better than WCP • Yields smaller queue and thus smaller delays • Not as good as optimal as we target 70% utilization 7 9 Optimal WCPCap TCP WCP

14. Simulations: Diamond Topology 1 3 • WCP does not achieve max-min rates • Rates are dependent on the number of congested neighborhood and the degree of congestion • WCPCap achieves max-min rates 2 4 5 6 8 7 9

15. Experimental Setup • Mini-PCs running Click and Linux 2.6.20 • ICOP eBox-3854 • 802.11b wireless cards running the madwifi driver • Omni directional antennas • some antennas covered with aluminum foils to reduce transmission range

16. Experimental Results: Stack Topology 1 3 2 4 5 6 8 7 9 Simulations Experiments For this topology, WCP’s simulation and experimental results are nearly identical

17. Experimental Results: Arbitrary Topology • 14 nodes and five flows • TCP starves different flows during different runs 18 18 23 23 10 10 24 24 16 16 22 22 12 12 26 26 13 13 15 15 19 19 11 11 20 20 14 14 WCP consistently gives fair rates

18. Related Work • WCP • Congestion control schemes explicitly recognizing neighborhood • NRED, EWCCP, and IFRC • Congestion control for ad-hoc wireless networks • TCP-F, TCP-ELFN, TCP-BuS, ATCP, etc. • COPAS, LRED, ATP, etc. • Congestion control for last-hop wireless networks • I-TCP, Snoop, WTCP, etc. • WCPCap • Heuristic based capacity estimation • WXCP and XCP-b • Schemes that also change the MAC layer • e.g, wGDP, DiffQ

19. Conclusions and Future Work • Demonstrate plausibility of distributed fair rate control for mesh networks • Low overhead AIMD scheme • Explicit rate feedback scheme • Future Work • Optimizing AIMD parameters in WCP • Reduce control overhead of WCPCap • More extensive experiments

20. Thank You