Journal of Computer Science

DETECTING MULTIPLE INTRUSION ATTACKS USING PERMANENT GIRTH CLUSTERING MODEL IN WIRELESS SENSOR NETWORK

G. Jayamurugan and P. Kamalakkannan

DOI : 10.3844/jcssp.2014.1473.1479

Journal of Computer Science

Volume 10, Issue 8

Pages 1473-1479

Abstract

Security is the central challenge and one of the serious concerns for designing reliable sensor networks. Of the different types of security threats in wireless sensor network, particularly dangerous attack is the replica node attack, in which the opponent takes the secret keying materials from a compromised node. It then produces large number of attacker-controlled replicas that divide up the cooperation node’s keying materials and ID. In this study we are specifically interested in investigating the extremely difficult problem concerning multiple attacks being routed in parallel with a given utility field and see if significant hypothetical solutions could be drawn. A clustering model for discovering multiple intrusions in WSN is identified. Permanent-girth Clustering (PC) model used to detect abnormal traffic patterns and then uses PC model to built the normal traffic behavior. In this study we develop mechanisms so that the PC model is capable to distinguish attacks. Furthermore, the detection scheme is based on position of traffic features that potentially are practical to an extensive variety of routing attacks. In order to approximate intrusion detection scheme, extensive sensor network simulator producing routing attacks in wireless sensor networks is designed. PC model for intrusion detection is capable to attain high detection accuracy with a low false positive rate for a multiplicity of replicated routing attacks. NS2 simulator is used to perform the experimental work of PC model on wireless sensor network. The experimental evaluation of PC model is measured in terms of average delay measurement, energy consumption and low false positive rate.

Copyright

© 2014 G. Jayamurugan and P. Kamalakkannan. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.