A Novel Method Based on Clustering Algorithm and SVM for Anomaly Intrusion Detection of Wireless Sensor Networks

Based on the principle that the same class is adjacent, an anomaly intrusion detection method based on K-means and Support Vector Machine (SVM) is presented. In order to overcome the disadvantage that k-means algorithm requires initializing parameters, this paper proposes an improved K-means algorithm with a strategy of adjustable parameters. According to the location of wireless sensor networks (WSN), we can obtain clustering results by applying improved K-means algorithm to WSN, and then SVM algorithm is applied to different clusters for anomaly intrusion detection. Simulation results show that the proposed method can detect abnormal behaviors efficiently and has high detection rate and low false positive rate than the current typical intrusion detection schemes of WSN.