A Congestion Avoidance Algorithm Based on Fuzzy Controlling

Many researches and experiments have shown that delay-based congestion avoidance (DCA) algorithms improve TCP throughputs. However, many factors affect the veracity of DCA algorithms. In order to solve these problems, a fuzzy control based on DCA algorithms (FDCA) is proposed which eliminates these negative influences when gauging network congestion conditions. The performance of the algorithm has been tested and evaluated using NS simulator and the simulation results have demonstrated that the algorithm identifies network congestions more accurately and consequently increases network throughputs.