Supporting QoS in wireless MAC by fuzzy control

An effective, efficient and distributed fuzzy control algorithm is proposed to perform control over the contention parameters in the IEEE 802.11 media access control (MAC) layer for supporting quality of service (QoS) in wireless communications. The proposed fuzzy control algorithms reserves bandwidths by controlling the delays of mobile nodes in the network, so that the bandwidths and other network resources can meet the constraints of different transmission classes. The proposed fuzzy control comes from generalized fuzzy automata (GFA) theory. To enhance the accuracy, the universe of control is also adapted according to network states. Furthermore, to reduce the computation cost of the controller, the paper also proposes a fuzzy mean approximation scheme that catches up delay variations and issues controls at adequate time instances. Therefore, the reference target delay can be quickly tracked, the settling time is enormously decreased, the bandwidth constraint is met, channel utilization is enhanced, and then the quality of the network is assured. In addition to proof of the stability property, we performed simulations to show that the proposed fuzzy control effectively and efficiently adapts the parameters of the contention window, such that the bandwidth specifications in the service agreement for mobile nodes are satisfied, even when the specifications are dynamic and the network background traffic has bursts.