Fuzzy Based Anomaly Intrusion Detection System for Clustered WSN

摘要

In Wireless Sensor Networks (WSN), the intrusion detection technique may result in increased computational cost, packet loss, performance degradation and so on. In order to overcome these issues, in this study, we propose a fuzzy based anomaly intrusion detection system for clustered WSN. Initially the cluster heads are selected based on the parameters such as link quality, residual energy and coverage. Then the anomaly intrusion is detected using fuzzy logic technique. This technique considers the parameters such as honest, energy level, unselfishness and prediction variance of each cluster member and provides the optimal trust threshold of the node as the result. By simulation result, we show that the proposed technique enhances the detection accuracy and reduces the false positive rate.