Malicious nodes detection algorithm based on triangle module fusion operator in wireless sensor networks

Abstract

The characteristics of wireless sensor networks and its deployment environment determine that wireless sensor networks is vulnerable to intrusion. In this paper, a malicious nodes detection algorithm based on triangle module fusion operator (MDTMO) is proposed for selective forwarding attack. The algorithm establishes the membership functions according to data packets that received and forwarded by the node, it uses the fusion method of triangle module operator to determine the suspected malicious node. Subsequently, the base station node detects the buffer occupancy and channel occupancy of the suspected malicious node to judge whether the network is in congestion. If the network quality is good, it is considered that the packets loss is due to the selective forwarding attack instead of congestion, so the node is judged to be a malicious node. The experimental results show that compared with the algorithm in [6], MDTMO has improved performance in both the malicious nodes detection rate and false positive rate.