Novel Intrusion Prevention and Detection Model in Wireless Sensor Network

Abstract

Abstract WSN have gotten progressively one of the most sultry exploration areas in the field of computer science because of their broad scope of uses, including military, transportation, and detection applications. To guarantee the security and reliability of WSN, an intrusion detection system ought to be set up. This IDS must be viable with the attributes of WSNs and fit for recognizing the most significant conceivable number of security dangers. This paper proposes an intrusion prevention and detection system in wireless sensor networks. The proposed model is a combination of FzMAI and Support Vector Machine. This technique performs in two phases: The first phase prevents intrusions from passing a wireless sensor network, and in the second phase, malicious nodes that joined the system are detected using Support Vector Machine. The proposed model is verified on WSN-DS dataset and yields an accuracy of 99.97%. The proposed model is compared with five existing algorithms: SVM, Naïve Bayes’, Random Forest, KNN, and Decision Tree and outperforms all five algorithms.KeywordsWireless sensor networksIntrusion detectionIntrusion preventionFzMAI