GENE POPULATED SPECTRAL CLUSTERING FOR ENERGY EFFICIENT MULTIPLE INTRUSION DETECTION AND RESPONSIVE MECHANISM FOR MANET

Abstract

A mobile ad hoc network (MANET) is a structure less network where the mobile devices are moved in random manner. In MANET, Each mobile device is randomly moves in various directions in the network. A few intrusions occurred due to the movement of mobile nodes in network. Mobile Nodes in an ad-hoc network are preserved by limited battery power for their operation. Hence, Energy management is a significant concern in a mobile ad-hoc network. In order to improve energy efficient multiple intrusion detection and responsive mechanism, Gene Populated Spectral Clustering (GPSC) technique is introduced in MANET. Initially, gene population generation is carried out to form a cluster. For each node, the energy and trust value is measured to detect the attacks. After that, various attacks such as grayhole, blackhole, wormhole, sleep deprivation and rushing attacks are identified through spectral clustering. The GPSC technique calculates the energy and trust value for each node in the cluster. Based on energy and trust value to select the cluster head and identify the intrusion levels. Finally, the intrusion response mechanism is performed from the intrusion level classifications. This helps to provide an efficient response with low network degradation. The severity of attack and the degradation in network performance provides efficient results from an implementation. Therefore, the proposed GPSC technique is more responsive to specific attack with spectral cluster. The simulation is