Cluster-based Cooperative Back Propagation Network Approach for Intrusion Detection in MANET

Mobile ad-hoc networks (MANET) are particularly vulnerable on account of its intrinsic characteristics of open medium, dynamic topology, absence of central authorities, distributed cooperation and constrained capability. These vulnerabilities create significant challenges for routing protocols operating in the entire network. In which, the reactive routing, i.e. AODV, bears the brunt of various kinds of attacks. In this paper, we try to build an efficient defense system based on a cooperative scheme to detect intrusions in AODV-based ad hoc networks using clustering technique and Back Propagation Network (BPN). A clustering architecture provides network scalability and fault tolerance, and results in more efficient use of network resources. Back-propagation neural networks is used for the purpose of anomaly detection and the feature is selected from the packets. The effectiveness of the proposed scheme is illustrated by means of extensive simulations using NS-2 simulator. Specifically, the comparison between BPN and finite state machine (FSM) is given.