Cluster Head Based Intrusion Detection System for Black Hole Attacks in Wireless Ad Hoc Networks using 2 Level Fuzzy Logic System

Ad hoc networks are autonomous and infrastructure-less wireless systems where nodes act as routers and hosts. Security is the primary issue for the functionality of these networks. Security for ad hoc networks can be incorporated by prevention and detection mechanisms. This research work focuses on a two-level fuzzy-based intrusion detection system for identifying black hole attacks in ad hoc networks. This method can reduce the complexity of the rule base of the fuzzy inference system. To reduce the complexity of detection, communication overhead and to make the detection scheme energy efficient, further, a cluster-head-based intrusion detection system is designed and implemented. The impact on network performance with no attack, with black hole attack, and with intrusion detection scheme deployed in all nodes and cluster heads are analyzed. The proposed cluster-based 2 level fuzzy logic intrusion detection mechanism was able to achieve the detection rate and accuracy to a maximum of 100%,false alarm rate to 0% and detection delay to in varying attacker scenario.