Intrusion Detection Based on Piecewise Fuzzy C-Means Clustering and Fuzzy Naïve Bayes Rule

Abstract

Intrusion detection has paramount importance in network security. Intrusion detection depends on energy dissipation, whereas trust remains a hectic factor. In this paper, a trust-aware scheme is proposed to detect intrusion in Mobile Ad Hoc Networking (MANET). The proposed method uses Piecewise Fuzzy C-Means Clustering (pifCM) and fuzzy Naive Bayes (fuzzy NB) for the intrusion detection in the network. The pifCM helps to determine the cluster heads from the clusters. After the selection of cluster heads, the intrusion in the network is determined using fuzzy Naive Bayes with the help of node trust table. The node trust table contains the updated trust factors of all the nodes and the presence of intruded nodes are found with the help of the trust table. After the intrusion is detected, they are eliminated and this reduces the delay in transmission. The effectiveness of the proposed method is analyzed based on the metrics, such as throughput, detection rate, delay, and energy. The proposed method has the delay at the rate of 0.003, energy dissipation of 0.657, the detection rate of 9.85, and throughput of 0.659.