Power-Aware Hybrid Intrusion Detection System Using Support Vector Machine in Wireless Ad Hoc Networks

Ad hoc wireless network with their changing topology and distributed nature are more prone to intruders. The network monitoring functionality should be in operation as long as the network exists with nil constraints. The efficiency of an Intrusion detection system in the case of an ad hoc network is not only determined by its dynamicity in monitoring but also in its flexibility in utilizing the available power in each of its nodes. In this paper we propose a hybrid intrusion detection system, based on a power level metric for potential ad hoc hosts, which is used to determine the duration for which a particular node can support a network-monitoring node.. Power -aware hybrid intrusion detection system focuses on the available power level in each of the nodes and determines the network monitors. Power awareness in the network results in maintaining power for network monitoring, with monitors changing often, since it is an iterative power-optimal solution to identify nodes for distributed agent-based intrusion detection. The advantage that this approach entails is the inherent flexibility it provides, by means of considering only fewer nodes for re-establishing network monitors. The detection of intrusions in the network is done with the help of support vector machine (SVM). The SVM's classify a packet routed through the network either as normal or an intrusion. The use of SVM's enable in the identification of already occurred intrusions as well as new intrusions.