WMCD: A Situation Aware Multicast Congestion Detection Scheme Using Support Vector Machines in MANETs

Congestion is one of the most important issues impeding the development and deployment of IP multicast and multicast application in Mobile ad-hoc network (MANETs). In this paper, we propose a situation aware multicast congestion detection scheme with support vector machines in MANETs. We focus on using support vector machines to detect incipient multicast congestion by using structural situation information. In this way, by using a situation aware learning system, we can detect incipient congestion in advance instead of waiting packet loss. The rate adaptation algorithm can reduce the transmission rate only if the loss is classified as a congestion loss. Simulation results show that a support vector machine is an appropriate mechanism for decision making in proactive multicast congestion detection.