A Statistical Study of Loss-Delay Tradeoff for RED Queues

Abstract

Abstract-Aside from the introduction of many new schemes, the use of TCP-based AQM schemes and in specific RED is anticipated to continue in foreseeable future as the de-facto standard of network congestion control. Therefore, conducting extra research work aiming at improving the performance of RED is still a topic of high interest. In this paper, we present an analytical study aiming at the fine tuning of the RED parameters. Utilizing a statistical analysis approach, we formulate an optimization problem aimed at addressing the loss and delay tradeoff of the RED queuing discipline. We provide a two-phase iterative solution to the problem in order to identify the settings of the RED parameters. We discuss the convergence characteristics of our solution and investigate its low complexity characteristics. Through extensive NS2 experiments, we illustrate the advantages of our proposed optimization approach by comparing its results to those of adaptive RED as well as standard RED with recommended parameter settings.Index Terms-RED, Markov chain queue modeling, optimal parameter fine tuning.

A New Approach to Estimate RED Parameters Using Function Regression

“…The packet delivery ratio obtained by the experiment results is in the range of our forecast. What is more, it is higher than that in literature [3]. This will provide us an approach to predict the network parameters according to the tolerance values of the packet delivery ratio, and then it makes the whole network process under control.…”

“…Finally, we compute the average of them running 20 simulation experiments in different pause time. The results are compared to AODV, AOMDV and literature [3], as shown in Figures 3, 4 Figure 3 shows that the average end-to-end delay is reduced both in the literature [3] and the mechanism proposed in this paper. However, our mechanism has smaller average end-toend delay when the pause time increases up to 100s and above.…”

“…Utilizing a statistical analysis approach, [3] formulated an optimization problem aiming at addressing the loss and delay tradeoff of RED queuing discipline. It is a two-phase iterative solution.…”

“…This will cause severe congestion of RREQ packets. To solve this problem, literature [3] drops all RREQ packets when the length of queue is longer than the threshold values. Our mechanism drops the arriving RREQ packets with a certain probability while increasing a little delay when the nodes move speedily.…”

“…A significant research work focused on the fine tuning of the RED parameters. Literature [3] presented an analytical study aiming at addressing the loss and delay tradeoff of the RED queuing discipline in order to identify the settings of the RED parameters. Literature [4] proposed a Statistical Adapting RED to dynamically tune RED parameters based on the standard deviation of instantaneous queue size.…”