A refined 3D Markov model for non-saturated IEEE 802.11 DCF netwoks

This paper proposes a novel analytical model to evaluate the system performance of the IEEE 802.11 MAC sublayer under both light and heavy traffic loads. This approach, based on a three-dimensional Markov chain, generalizes the bi-dimensional Markov chain models introduced in the literature. Traditional bi-dimensional Markov chain models, while suitable for throughput analysis, are unable to capture QoS performance metrics such as the average packet delay, the average queue length and the blocking and discard probabilities due to the lack of a proper queueing model. We present a refined three-dimensional Markov chain approach that incorporates a queueing model and error prone channel effects. The Markov chain is efficiently solved by using our refined Collapsed Transition onto Basis approach, allowing the investigation of the whole set of QoS metrics for an IEEE 802.11 network under realistic conditions. The model is validated by contrasting the predicted metrics with extensive numerical simulations as well as with results obtained using previously available analytical methods.