HYBRID INTRUSION DETECTION SYSTEM FOR WIRELESS SENSOR NETWORKS

Abstract

In recent years there is a rapid growth in the field of sensor networks. There are several applications including environmental monitoring, military, rescue and smart homes. Basically WSN composed of many tiny sensors which communicate with each other wirelessly. Sensor nodes mainly perform three tasks sensing, processing and communication .Sensor nodes are generally deployed in open and hostile environment thus they can be attacked easily. Wireless sensor networks are exposed to different types of security threats that can reduce the performance of the whole network; that might result in serious problems like denial of service attacks, routing attacks, Sybil attack etc. Defensive mechanisms like Key management protocols, authentication protocols and secure routing cannot provide security to WSNs for these types of attacks. Intrusion detection system is a solution to this problem. It analyses the network by collecting sufficient amount of data and detects anomalous behavior of sensor node. IDS based security mechanisms suggested for other network models such as ad hoc networks, cannot directly be used in WSNs. Researchers have proposed various intrusion detection systems for wireless sensor networks during the last few years. In this dissertation, a simple hybrid intrusion detection system (IDS) for wireless sensor networks is designed and implemented. The proposed IDS use both of anomaly and misuse detection. In order to get hybrid system, combination of anomaly and misuse detection techniques. These techniques provide high detection rate and high accuracy of detection. In addition, a cluster-based wireless sensor network is used to reduce communication costs and packet overheads.