A Novel Model of Sybil Attack in Cluster-Based Wireless Sensor Networks and Propose a Distributed Algorithm to Defend It

Abstract

Today, Wireless Sensor Networks are widely employed in various applications including military, environment, medical and urban applications. Thus, security establishment in such networks is of great importance. One of the dangerous attacks against these networks is Sybil attack. In this attack, malicious node propagates multiple fake identities simultaneously which affects routing protocols and many other operations like voting, reputation evaluation, and data aggregation. In this paper, first, a novel model of Sybil attack in cluster-based sensor networks is proposed. In the proposed attack model, a malicious node uses each of its Sybil identity to join each cluster in the network. Thus, the malicious node joins many clusters of the network simultaneously. In this paper, also a distributed algorithm based on Received Signal Strength Indicator and positioning using three points to defend against the novel attack model is proposed. The proposed algorithm is implemented and its efficiency in terms of true detection rate, false detection rate, and communication overhead is evaluated through a series of experiments. Experiment results show that the proposed algorithm is able to detect 99.8% of Sybil nodes with 0.008% false detection rate (in average). Additionally, the proposed algorithm is compared with other algorithms in terms of true detection rate and false detection rate which shows that the proposed algorithm performs desirably.