An Improved Clustering-Based Approach for Wormhole Attack Detection in MANET

Abstract

Advancement in wireless technologies and improved use of wireless devices demand more and more infrastructure less networks like Mobile Ad hoc Networks (MANET). MANETs are becoming common, since they can be deployed easily in a cost effective manner wherever needed. The high usage of MANET demands more security and confidentiality of the data communicated through MANET. These networks are inherently vulnerable to various attacks due to the absence of concealed channel. Wormhole attack is a great threat to MANET, as it is capable of launching several other types of attacks like black-hole attack, gray-hole attack, sink-hole attack, DDoS, etc. A strategically placed wormhole node can disrupt the entire communication in the network. In this paper, wormhole attack launched by exploiting AODV protocol in MANET, is detected and eliminated in two phases. The preliminary phase in the process of identifying wormhole attack is done, based on timing analysis and hop count. After suspecting the attack, a Clustering based approach is used to confirm the presence of attack, and also to identify the attacker nodes. The entire network is divided into different clusters and each cluster will have a Cluster Head, which controls all the nodes in the cluster and plays the role of a controlling authority in MANET.