The Application Research Of PID Active Queue Management Algorithm In Congestion Control

With the popularity of Internet, the demand for network quality of service(QoS) is higher and higher. Effective congestion control is one of the most important factors. During the early time of designing the Internet, people usually implemented it by means of end-to-end flow control scheme based on slide window in Transmission Control Protocol(TCP). Flow control plays an important role in guaranteeing the stability of the network. The recent research begins to explore how to make the intermediate node contribute to congestion avoidance because there are some limitations at end system.As an enhancement mechanism for the end-to-end congestion control, Active Queue Management(AQM) can keep smaller queue delay and higher throughput by purposefully dropping the packets at the intermediate nodes. So it can guarantee the QoS.In this paper a linearized model of TCP and AQM firstly is presented. Then we analyze the Random Early Detection(RED) algorithm using ns-2 simulations and point out that it is difficult in tuning the parameters because the algorithm is instinctive and heuristic. In order to solve this problem, in this paper , after analyzing the P and PI controller, we apply the control theory to convert the design of AQM algorithm to the design of the PID controller. Firstly we design a good continuousPID controller using the classically analytical method in frequency domain; Secondly we discretize it by "Bilinear Transformation" and get the digital controller; Finally we implement it in NS-2. The simulation results verify that the performance of the PID controller are better than those of "RED" and PI controller at many aspects.