We derive a complete characterization (marginal distribution, autocovariance) of the service time process in a saturated IEEE 802.11 wireless LAN, both from the point of view of a single station and of the system as a whole.
Our service time distribution model is then exploited to highlight the burstiness of service times and its dependence on the maximum value of the IEEE 802.11 contention window. We discuss a trade-off between throughput efficiency and service time variance, showing that a minor throughput loss can bring about a major benefit in service time smoothness, at least for practical values of the number of simultaneously competing stations (e.g.\ less than 15).