A MODIFIED ROBUST ADAPTIVE CONGESTION CONTROLLER FOR ABR TRAFFIC

Abstract

In the computer networks, congestion occurs when the resource demands exceeds the capacity, which the network can provide. The presence of too many packets in the network, which the network is not able to cope with due to its limited resources and it begins losing packets, gives birth to the situation called congestion. ATM networks have to deal with the two types of congestions, first the long-term congestion, caused by more traffic coming in than the system can handle, and second the short-term congestion caused by burstiness in the traffic. The aims of this dissertation is to analyze the performance of Robust Adaptive Congestion Control algorithm for ABR traffic in ATM networks and modify this algorithm for better performance. The main goal of ABR traffic control is to provide fairness among all the VCs with a minimal cell loss ratio and maximum utilization of the network. A simulation framework is developed in C++ programming language on which both (previous and modified) algorithms are implemented and the simulation results for performance thus obtained are presented. The results show that both the algorithms achieve basic fairness and max-min fairness but the modified algorithm reduces the settling time.