A Novel Approach for Efficient Usage of Intrusion Detection System in Mobile Ad Hoc Networks

Abstract

Mobile ad hoc networks (MANETs) are self-configuring infrastructureless dynamic wireless networks in which the nodes are resource constrained. Intrusion detection systems (IDSs) are used in MANETs to monitor activities to detect any intrusion in the otherwise vulnerable network. In this paper, we present efficient schemes for analyzing and optimizing the time duration for which the IDSs need to remain active in a MANET. A probabilistic model is proposed, which makes use of cooperation between IDSs among neighborhood nodes to reduce their individual active time. Typically, an IDS has to run all the time on every node to oversee network behavior. This can turn out to be a costly overhead for a battery-powered mobile device in terms of power and computational resources. Hence, in this work, our aim is to reduce the duration of active time of the IDSs without compromising their effectiveness. To validate our proposed approach, we model the interactions between IDSs as a multiplayer cooperative game in which the players have partially cooperative and partially conflicting goals. We theoretically analyze this game and support it with simulation results.