Cognitive fuzzy flow control for wireless routers

This paper presents fuzzy set theory-based cognitive control system for IEEE 802.11b wireless local area networks (WLANs). Developed fuzzy weighted queueing (FWQ) method anticipates required changes on weight coefficients with optimal packet sizes for adaptive flow control. Traffic flows are scheduled for prevailing traffic level on WLAN-based router. The algorithm determines the amount of allowed bandwidth for each service class in the outputs of wireless router anticipating the application dependent delay and packet loss rate. It is shown through simulations that the developed FWQ model is also more stable and reacts faster to different traffic states than drop-tail or weighted fair queueing (WFQ) schedulers that were used here as comparative methods. Delay times and packet loss rates of the FWQ algorithm were lower than drop-tail's or WFQ's respective values with different amounts of background traffic.