Normalized throughput and delay analysis using QMBCCA in wireless environment

Abstract

Network Congestion is a key problem in today's Internet. TCP is used as a Transport Protocol for most of the applications on the Internet and is also responsible for adjusting to network congestion dynamically. TCPs congestion control mechanism is insufficient in a wireless environment because packet loss does not always mean congestion, but can be due to the sudden drop in channel quality. The throttling measures of TCP often result in an unnecessary reduction of end-to-end throughput which results in suboptimal performance. Hence the sender has to be made aware that some of the losses reported are not due to congestion. The Active Queue Management algorithms (AQM) are used to reduce congestion, and in this paper three such AQM algorithms are considered. We look at the existing Random Early Detection (RED), Wireless Explicit Congestion Notification (WECN) algorithm and an algorithm proposed by us called as the Queue Management Backward Congestion Control Algorithm (QMBCCA). A comparison of the three algorithms in terms of Normalized Throughput and Delay in a Wireless Environment is done in this paper. We have found that the Normalized Throughput is always more, and Delay is always less with respect to QMBCCA when compared with the other two algorithms.