Analytical Modeling of IEEE 802.11 Type CSMA/CA Networks with Short Term Unfairness

TL;DR

This paper develops an accurate stochastic model for IEEE 802.11 DCF networks with non-standard parameters and propagation delays, revealing short term unfairness issues and enabling better protocol tuning.

Contribution

It introduces a novel approximate analytical method that accurately predicts short term unfairness in IEEE 802.11 networks with large delays and parameter variations.

Findings

Identifies conditions leading to short term unfairness.

Provides a computationally feasible model for performance prediction.

Enables protocol parameter tuning for fairness and throughput.

Abstract

We consider single-hop topologies with saturated transmitting nodes, using IEEE~802.11 DCF for medium access. However, unlike the conventional WiFi, we study systems where one or more of the protocol parameters are different from the standard, and/or where the propagation delays among the nodes are not negligible compared to the duration of a backoff slot. We observe that for several classes of protocol parameters, and for large propagation delays, such systems exhibit a certain performance anomaly known as short term unfairness, which may lead to severe performance degradation. The standard fixed point analysis technique (and its simple extensions) do not predict the system behavior well in such cases; a mean field model based asymptotic approach also is not adequate to predict the performance for networks of practical sizes in such cases. We provide a detailed stochastic model that accurately captures the system evolution. Since an exact analysis of this model is computationally intractable, we develop a novel approximate, but accurate, analysis that uses a parsimonious state representation for computational tractability. Apart from providing insights into the system behavior, the analytical method is also able to quantify the extent of short term unfairness in the system, and can therefore be used for tuning the protocol parameters to achieve desired throughput and fairness objectives.