Analysis for protocols with self similar traffic

IEEE 802.11 wireless local area network (WLAN) is one of the most deployed wireless technologies all over the world. An accurate and comprehensive performance evaluation is crucial for the quality of service (QoS) guarantee in 802.11 networks. In this paper, we use the method of stochastic network calculus to obtain the stochastic delay bounds of 802.11 protocols. In particular, we take the self-similar property of wireless traffic into consideration and fractional Brownian motion (fBm) traffic model is adopted to describe the arrival traffic.

We derive the stochastic arrival curve of fBm traffic and the stochastic service curve of existing model of 802.11, then obtain the stochastic delay and backlog bounds. Finally, we use matlab to calculate theoretical results and ns-2 for simulations. The results show that the delay of the 802.11 wireless node increases with the increasing of Hurst parameter H and the number of competition stations n. And the results also indicate that the theoretical results well bound simulation results.