Modified Fuzzy-Based Random Early Detection for Constant Bit Rate and Poisson Traffic

The Random Early Detection (RED) mechanism, attempts to control the congestion level at a bottleneck by monitoring and updating the average queue size. The RED algorithm calculates the average queue size using a low pass filter with an exponential weighted moving average. The average queue size is compared to two thresholds: a minimum and a maximum threshold. However in this technique, the probability of dropping the packets is more which leads to data loss. This paper proposes a modified fuzzy based RED algorithm with more levels of thresholding to effectively avoid congestion. It also minimizes data loss. Keywords: Active Queue Management (AQM), Random Early Detection (RED), Gentle RED, Piecewise RED, Threshold, Fuzzy .