Kalman Filtering: Estimate of the numbers of active queues in an 802.11e EDCA WLAN

Many schemes that aim to tune the IEEE 802.11e EDCA transmission parameters have been proposed in the literature. Based on them, we notice that the knowledge of the numbers of active transmission queues of each Access Category has an important role in modeling and optimizing the network. In this paper, we propose a simple and standard-compliant mechanism that processes easily obtained traffic measurements into a novel Extended Kalman Filter to obtain estimates of those numbers. For validating the filter, we simulate three network scenarios. The first two comprehend an application of the filter under saturated traffic and the third one refers to non-saturated conditions. Assessing the network performance impact promoted by the application and analyzing the EKF behavior in diverse traffic conditions we find out that under saturation the filter is precise and accurate enough to match closely the ideal performance and under non-saturated traffic the estimates track the short-term average numbers of active queues.