A network calculus approach to delay evaluation of IEEE 802.11 DCF

Abstract

Stochastic network calculus is an evolving theory for network performance guarantee analysis. Although many theoretical results of this theory have been developed, there still lack applicable examples to demonstrate how it may be used. This paper exemplifies applying stochastic network calculus to delay analysis of the IEEE 802.11 distributed coordination function (DCF). Analyzing and obtaining the stochastic characteristics of the single packet service time is the primary task. Then the stochastic behavior of the DCF is characterized by a time-domain server model, which describes the cumulative service time provided to an arrival flow using a probabilistic bound. Based on this server model, we obtain delay bounds for different arrival processes. In addition, the delay bounds also take buffer size into account. The analytical bounds are further discussed using numerical results. Through these, we present a stochastic network calculus approach to delay evaluation of the DCF.