Wireless unfairness: alleviate MAC congestion first!

Authors: Kan Cai, Michael Blackstock, Reza Lotun, Michael J. Feeley, Charles Krasic, Junfang Wang University of Columbia, University of Cincinnati

Complete Citation

Cai, K., Blackstock, M., Lotun, R., Feeley, M. J., Krasic, C., and Wang, J. 2007. Wireless unfairness: alleviate MAC congestion first!. In Proceedings of the the Second ACM international Workshop on Wireless Network Testbeds, Experimental Evaluation and Characterization (Montreal, Quebec, Canada, September 10 - 10, 2007). WinTECH? '07. ACM Press, New York, NY, 43-50. DOI= http://doi.acm.org/10.1145/1287767.1287777

Abstract

It is well known that competition between 802.11 wireless flows can lead to severe unfairness problems. In this paper we show that certain flows can be denied from their fair share of network access for long periods of time due toinequitable channel conditions. We argue that this unfairness problem is likely to get worse as the use of wireless in terms of number of devices and bandwidth continues to grow. To address unfairness, we rely on a common network management technique called traffic shaping. Instead of designing new MAC protocols or introducing complex wireless fair queuing or adaptation algorithms, we argue that 802.11, TCP and a well-understood wired fair queuing scheme can provide fairness even in unfavourable network topologies as long as we can prevent MAC-layer congestion from happening. This shaping approach has the advantage of being easily deployed in existing network management systems, requiring no change to the end-user devices or access points. We have proven the effectiveness of this cross-layer approach using a wireless testbed, and are currently in the process of incorporating this scheme into a large-scale campus wireless network.

Annotations

This paper is focused on an providing fairness in an infrastructure based wireless network. The basic premise is that all traffic should go through a router that is hardwired to all of the wireless access points. From there the router will perform traffic shaping and fair queuing based upon the destination AP and what other APs they are likely to interfere with. Thus, the shaper then tries to ensure that each flow gets the needed bandwidth to maintain fairness across all of the wireless nodes. In their experimental studies, (1 room a handful of routers/wireless nodes), they are able to ensure fairness between flows as long as the total bandwidth consumes by the flows is less the 20Mbps (which seems to be the maximum attainable bandwidth utilization for their network). After the aggregate bandwidth surpasses 20Mbps the flows are no longer able to maintain fairness.

The thing that struck me as interesting about this paper is their definition of fairness. Meaning, they believed that a node that is operating under poor conditions should have a similar goodput to competing flows that are not being interfered with. I think this is a mistake, because if you have a single device malfunction that causes one device not to receive 80% of its packets or something, the whole network will be brought to it's knees as the shaper allocates more and more bandwidth so that the lossy node can achieve a higher level of goodput.

-- DavidSalyers - 03 Oct 2007