Modeling of IEEE 802.11 based Multi-rate Wireless Ad Hoc Networks and Its Application towards Optimal Network Operation

In this paper, we derive an analytical model to evaluate the effect of jointly tuning the carrier sense threshold and employing multiple data rates on the network capacity in multi-rate, multi-hop wireless networks. The proposed model characterizes the channel status within the carrier sense range of an individual transmitter. The per-node throughput attained by each transmitter is then obtained by deriving the probability that each state occurs and its expected duration. The effect of multiple channel rates provided by IEEE 802.11 a/b/g protocols is also taken into account. As a preamble to devising a localized algorithm for tuning systems parameters, we propose a control reference and derive, for each data rate, its optimal value, the range of which remains approximately constant regardless of network dynamics. Based on the insight shed from the analytical model, we then devise a distributed and localized algorithm with which each node first selects a data rate and the associated carrier sense threshold based on the network condition, and then tunes the contention window size so that the value of the control reference measured stays in its optimal range. We show via simulation that the proposed algorithm, as compared to the existing IEEE 802.11 DCF protocol, yields a higher network capacity.