131 An Attack Identification Scheme Using Hierarchical Data Clustering in MANET

The nature of mobile computing environment in MANET makes it severely vulnerable to the malicious attacks by intruders. A compromised node can easily launch a denial of service attack by either dropping packets, as in blackhole attack, or by flooding the network with a large number of RREQs to invalid destinations in the network, as in flooding attack, thus trying to sabotage the other nodes or even the whole network. To combat with these adversaries, it is necessary to deploy an intrusion detection and response scheme. IDS with anomaly detection techniques use data mining to capture audit data and reason with the evidence in that data to determine whether the system is under attack. In this paper, a hierarchical data clustering scheme, using data mining for anomaly detection, is proposed for identifying the blackhole and flooding attacks in MANET, by observing the network packets.