A Predictive and Robust Active Queue Management for Internet Congestion Control

  • Authors:
  • Affiliations:
  • Venue:
  • ISCC '03 Proceedings of the Eighth IEEE International Symposium on Computers and Communications
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently many active queue management (AQM) algorithmshave been proposed to address performance degradationsof end-to-end congestion control. However, theseAQM algorithms show weaknesses to detect and controlcongestion under dynamically changing network situations.In this paper, we propose a predictive and robust AQMalgorithm, called proportional-integral-derivative (PID)-controller,using PID feedback control to remedy the weaknessesof existing AQM proposals. The PID-controller isable to detect and control the incipient as well as currentcongestion adaptively and proactively to dynamicallychanging network environments. A simulation study overa wide range of IP traffic conditions shows that PID-controlleroutperforms other AQM algorithms such as RandomEarly Detection (RED) [9] and Proportional-Integral(PI) controller [13] in terms of the queue length dynamics,the packet loss rates, and the link utilization.