A novel method for queqe managment using RED technique in mobile ad hoc network

Abstract

Mobile ad hoc network (MANET) consist of various communicating nodes & networking devices without having fixed topology/infrastructure. These components of MANET usually communicate with each other through wireless medium of communication. In this arrangement i.e. in MANET the congestion occurred at any intermediate node results in Long delays & packet loss which ultimately degrades the overall performance of whole network. One way of controlling this congestion is by efficient Active Queue Management [AQM] like Random Early Detection [RED]. Using RED states a lot of uncertainty because it highly depends on fixed parameters values which derives average queue length. Adjusting these parameters according to the current traffic load becomes extremely important. A network can works at its best if queue length of its nodes is adjusted according to the changes in current traffic load. This paper presents an algorithm which adjusts its threshold parameters & packet drop probability according to the current traffic load. The proposed approach will follow adaptive mechanism to adjust the RED Queue thresholds parameters for efficiently managing average queue length. The main idea is that as traffic load changes and the queue length will length will be changed accordingly, this algorithm will for sure enhance the network performance over the traditional RED technique.