Dynamic spatial backoff in fading environments

We present a dynamic spatial backoff method to resolve channel contention in wireless ad-hoc networks. We argue that each node should adjust its receiver sensitivity level according to the mean channel gain of its particular transmitter-receiver link, in order to see the full benefit of spatial backoff and improve the throughput. We designed a distributed algorithm that adjusts each transmitters carrier sense threshold and transmission rate dynamically based on local information and limited receiver feedback, for wireless channels that have small-scale multipath fading. We evaluated the algorithm using different topologies, under various fading conditions. Results show that our algorithm is able to achieve aggregate throughput near or better than the maximum achievable by the static scheme, without a priori knowledge of the network topology or fading condition.