Accumulation based Congestion Control using Active Queue Management

A network is very important for fast communication in real life but it is suffered from many problems and one of them is packets loss due to congestion. Congestion Control helps to prevent internet from gridlock state. When a packets switch becomes congested, its buffer can overflow and packets loss can occur. The internet avoids this problem by forcing end system to decrease the rate of sending packets into the network during congestion period. The end system uses algorithm to control the congestion is called Congestion Control (CC) algorithm. The CC algorithm has three major components additive-increase, multiplicative-decrease and slow-start. Accumulation-based congestion control (ACC) has two mechanism, first is TCP Vegas which is used to control the congestion by estimating the accumulation at sender end. Another is, Monaco scheme that solve problems of TCP Vegas by employing an out-of-band, receiver-based accumulation estimator, with the support of two FIFO priority queues. These schemes suffer from router buffer scalability problem which is solved by the use of Active Queue Management (AQM) .This paper illustrates the AQM scheme on the basis of throughput, fairness, and scalability and these parameters are analyzed by Network Simulator (NS-2) using dumb-bell topology, single bottleneck.