Active Queue Management Using Non-Linear Adaptive Random Early Detection Mechanism

This thesis proposes a new Active Queue Management(AQM) algorithm called POWer Adaptive Random Early Detection (POWARED) for the Internet congestion control. POWARED uses non-linear control function to adaptively maintain the average queue size of router's buffer at a specific target queue size...

Full description

Saved in:
Bibliographic Details
Main Author: Ng, Boon Kiong
Format: Thesis
Published: 2006
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This thesis proposes a new Active Queue Management(AQM) algorithm called POWer Adaptive Random Early Detection (POWARED) for the Internet congestion control. POWARED uses non-linear control function to adaptively maintain the average queue size of router's buffer at a specific target queue size in which the queue system achieves its steady-state condition. By consistently driving the average queue size of the buffer to oscillate around the system's equilibrium operating point, the performance of the overall network is enhanced. Based on simulation results, we observed significant improvement in terms of link utilization , packet loss rates and average delay for the router deploying POWARED compared to the router using traditional Drop-Tail or other existing AQM algorithms.