Research On Active Queue Management Mechanism Based On Aggregated Flows And Complex Topologies

Now,it is widely used in Internet that AIMD(Additive Increase Multiplicative Decrease) congestion control mechanism of TCP(Transmission Control Protocol) protocol.It is called end-to-end congestion control that the mechanism depended on end nodes.The simple drop tail queue management in routers used by traditional end-to-end congestion control mechanism can lead to many problems,such as long delay,high packet loss rate,global synchronization,fairness between flows etc.To improve the performance of end-to-end congestion control as well as provider low delay,low loss and fair service in best-effort IP networks,IETF(Internet Engineering Task Force) suggested using Active Queue Management(AQM) in routers.In this thesis,we first describe many AQM algorithms and TCP protocols in recent years.We analyze the fluid-based model of TCP/AQM,and develop a linearization processing for fluid-based model at equilibrium points for the stability analysis of the networks with AQM.The bottleneck networks are described by uncertain linear time-delay systems,2-D Laplace-z transform has been applied in the stability test of the network.We analyze active queue management mechanism based on aggregated flows. It is verified by simulations.We analyze performance and stability of active queue management mechanism based on complex topologies.At the end of this thesis,we draw the conclusion and future research.