Congestion is an important issue which researchers focus on in theTransmission Control Protocol (TCP) network environment. To keep the stabilityof the whole network, congestion control algorithms have been extensivelystudied. Queue management method employed by the routers is one of theimportant issues in the congestion control study. Active queue management (AQM)has been proposed as a router-based mechanism for early detection of congestioninside the network. In this paper we analyzed several active queue managementalgorithms with respect to their abilities of maintaining high resourceutilization, identifying and restricting disproportionate bandwidth usage, andtheir deployment complexity. We compare the performance of FRED, BLUE, SFB, andCHOKe based on simulation results, using RED and Drop Tail as the evaluationbaseline. The characteristics of different algorithms are also discussed andcompared. Simulation is done by using Network Simulator(NS2) and the graphs aredrawn using X- graph.
展开▼