Fuzzy Two Time-scale Congestion Control Algorithm

Researches have shown that self similar traffic has essential effects on the network parameters such as loss rate, length of the queue in the routers and throughput. Since classical congestion control algorithms work in short time scales and do not consider self similarity and long time dependency phenomenon of data, development of congestion control algorithms which be able to distinguish and manage of such traffic have been very important to researchers. Multiple time-scale congestion control is a method which has high performance reliance co-efficiency. In this research congestion control in large time scales has been implemented in fuzzy form. Simulation results show that fuzzy congestion controller will have better performance in comparison with a non-fuzzy two time-scale controller, if fuzzy rules are selected correctly.