Analysis of a novel predictable RED Algorithm with Self-Similar Traffic Input
Hu Yin · Microcomputer Information · 2008
In this paper,an attempt has been made by proposing a novel predictable RED Algorithm(SSRED) for traffic congestion management in IP networks with self-similar input.The self-similar parameter H had been introduced into the weight function,and we had also derived the thresholds of the maximum and minimum queue size based on self-similar traffic modelling.Finally we had also modified the maximum packet dropping probability of the proposed algorithm according to buffer overflow probability that Norros had proposed.Simulation results demonstrate that this algorithm can be used to improve the utiliazation,reduce the packet loss rate and the network delay under the self-similar traffic.