Abstract
The node replication attack is considered one of the most dangerous attacks against wireless sensor networks (WSNs). In this attack, an adversary captures one or more normal nodes of the network, extracts its key materials, generates several replicas, and deploys them in the network. In this paper, we propose a novel, intelligent, and lightweight algorithm using learning agents and watchdog nodes (WNs) to detect replica nodes in mobile WSNs. In the proposed algorithm, there are a few WNs, each one equipped with a learning agent that monitors the network traffic and nodes’ movements to identify potential replica nodes in the network. We use the fact that the number of times that a replica node meets a WN is probably more than that of a normal node during a specific monitoring interval of the network. The proposed algorithm is simulated using the J-SIM simulator, and its performance is evaluated in terms of replica node–detection probability and false detection probability through experiments. Experiment results show that the proposed algorithm can detect 100% of the replica nodes, while its false detection probability is less than 0.01