We address the problem of identifying misbehaving nodes that (selectively) drop packets, in order to degrade the network performance. Such nodes may agree to forward packets by participating in the route discovery process, but refuse to do so once the packets have been received. We propose a reactive approach where the source initiates an audit process if a significant performance degradation is observed. We employ a compact representation of the behavioral proof of a node by adopting Bloom filter structures and show that the misbehaving node can be identified based on random audits. Our approach provides significant energy savings compared to previously proposed methods that rely on reputation systems, or intensive acknowledgment schemes.
Abstract. We address the problem of identifying misbehaving nodes that refuse to forward packets in wireless multi-hop networks. We map the process of locating the misbehaving nodes to the classic Rényi-Ulam game of 20 questions. Compared to previous methods, our mapping allows the evaluation of node behavior on a per-packet basis, without the need for energy-expensive overhearing techniques or intensive acknowledgment schemes. Furthermore, it copes with colluding adversaries that coordinate their behavioral patterns to avoid identification and frame honest nodes. We show via simulations that our algorithms reduce the communication overhead for identifying misbehaving nodes by at least one order of magnitude compared to other methods, while increasing the identification delay logarithmically with the path size.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.