A detection technique for identity based attacks in clustered mobile ad-hoc networks

Decentralized networks such as mobile ad-hoc networks (MANETs) are susceptible to many types of security attacks. Ideally, each node in a MANET should be associated with only one identity, but such networks are not equipped with a mechanism that can check whether one physical device is bound to exactly one identity or more. As a consequence, a malicious node can easily obtain more than one identity and behave as many independent nodes in the network. Such an attack comes under the category of identity based attacks and has a potential to disrupt the normal functioning of the network in many ways. In this paper, a detection mechanism has been proposed to find out the attacker which fools the other nodes in the network by using more than one identity at a time. The detection approach is based on the fact that all the illegitimately acquired identities of an attacker travel together on a single physical device.