A Robust Trust Model for Wireless Sensor Networks

Abstract

Wireless Sensor Networks (WSNs) are vulnerable to various security attacks due to its broadcast nature. Trust management is one of the effective methods to defend against threats posed to WSN. The trust management system helps to mitigate many insider attacks and malicious behaviour of nodes. In this paper, we proposed a robust trust model that prevents various attacks viz; bad mouthing, self-promoting and collusion attacks. It uses the time lapses function based on forgetting curve to calculate direct trust and reputation function for indirect trust. Our proposed trust model is scalable. When the trust value is above the threshold value, then we consider a communication link as active and node is trusted, otherwise we remove the communication link and mark the node as malicious. The simulation results indicate the effectiveness of proposed method and maximizes the defence against the internal attacks.