Issues in Wardrop Routing in Wireless Networks

Abstract

Traditionally, traffic in wireless networks is routed along minimum hop paths from sources to destinations. It can be theoretically shown that flow avoiding routing provides throughput gains over shortest path routing in random wireless networks by a factor of four when the sources are few enough. Motivated by this, an alternative approach to wireless routing using delay measurements to adaptively route packets in order to achieve a Wardrop equilibrium is presented. The proposed protocol is completely distributed and balances load along loop-free paths. An ns-2 simulation study indicates that the protocol is able to automatically route flows to avoid each other, improving over the performance of shortest-path protocols in a variety of scenarios. The architectural challenges in implementing such a multi-path, delay based, probabilistic algorithm are discussed. A user-space implementation of the protocol has been built on a modified Linux 2.4.20 kernel. Finally, a measurement study of the implementation on a six node testbed is discussed.