Comparison of Red and E-Red:A Stabilizing AQM Scheme for Congestion Control and Avoidance

In today's globally integrated Emerging technologies, the current Internet protocols uses packet-level mechanisms having end to end links to detect and react to congestion. But the regulation of best-effort traffic and congestion control is largely left to the Queue Management. Drop-tail, Random Early Detection (RED) and Random Exponential Marking are quite effective in reducing packet losses under certain level of congestion and are insufficient to ensure overall network stability. At times of high congestion, not only does the end user experience very poor performance, but also a lot of network resource is wasted due to retransmissions and aborted connections. Hence, it is required to propose a new efficient AQM (Active Queue Management) scheme known as Exponential Random Early Detection (E-RED) to reduce the packet loss even at more congestion time. The proposed probability calculation approach for congestion control in an ERED framework is implemented with probability marking capabilities. We had implemented ERED algorithm and compared with the RED algorithm during packet loss in congestion control at various time intervals. The experimental result shows that our proposed approach E-RED outperforms RED. The experimental result also shows the packet loss for the nodes at various time intervals, Queue sizes and bandwidth.