Adaptive Internet congestion control.

摘要

Currently proposed active queue management (AQM) algorithms detect and control congestion reactively based on current and/or past congestion. In this dissertation, we argue that AQM based transmission control protocol (TCP) congestion control should be adaptive to dynamically changing traffic situations. We outline requirements for an AQM algorithm to be adaptive to dynamically changing traffic situations and proactive in its attempts to control congestion.; We design two adaptive and proactive AQM algorithms, PID-controller and Pro-Active Queue Management (PAQM), using classical Proportional-Integral-Derivative (PID) feedback control to overcome the reactive congestion control dynamics of existing AQM algorithms. We first introduce and analyze a feedback control model of TCP/AQM dynamics, and use this to build a discretized control implementation of a PID-controller and PAQM method. PID-controller and PAQM can provide proactive congestion avoidance and control using an adaptive congestion indicator and control function for a wide range of traffic environments. PID-controller and PAQM stabilize the queue length around a desired level while giving smooth and low packet loss rates independent of the traffic load. A simulation study with a wide range of realistic traffic conditions suggests that PID-controller and PAQM outperform other AQM algorithms such as Random Early Detection (RED) and Proportional-Integral (PI) controller.