Unified fixed-point analysis of IEEE 802.11 WLAN under saturated and unsaturated conditions

Most of analysis so far for IEEE 802.11 wireless local area networks (WLANs) focuses on saturated condition. However, it is of practical value to take into account the unsaturation case. In this paper, we modified Bianchi's Markov back-off model to make it applicable to unsaturated condition and the analytic results are provided by employing the renewal-reward theorem. Under our proposed model, we study the fixed-point solution of the system and provide a condition to guarantee both the uniqueness and balance of the fixed point. From the fixed point, we find that under unsaturated condition, network parameters should be adjusted according to the traffic load. Then, we study the system throughput. In the case where there are a large number of nodes, we provide closed-form formulas for the collision probability, the aggregate attempt rate, and the throughput. We find that in such a scenario, the system yields similar performance as that under saturated situation. Moreover, we compare all the results with those under saturated condition and find the latter is a special case of our results. Hence, all of our analysis based on unsaturated condition well covers saturated condition. Our analytical results are validated through ns2 simulations. Copyright © 2010 John Wiley & Sons, Ltd.