Intrusion Detection using Information Gain Self Adaptive Elliptic Key in MANET Musthafa A. Syed*, Dr. Babu C. Nelson Kennedy** *Assistant Professor, Department of IT, K. S. Rangasamy College of Technology, Tamilnadu, India **Professor, Department of CSE, Dhanalakshmi Srinivasan College of Engineering, Tamilnadu, India Online published on 14 October, 2016. Abstract In recent years, Mobile Ad-hoc Networks (MANETs) have received massive importance in various fields due to their dynamic networking services. Such networks are quickly adaptable in the future hence intrusion detection remains the key for a secure ad hoc environment. However, most of these systems either use pre-distributed keys or compromises the detection rate by applying rational model. In this paper, an intrusion detection method, called, Information Gain Self Adaptive Elliptic Key Generation (IG-SAEKG) in MANET is presented to improve the detection rate and increase the packet delivery rate with minimum delay. For this purpose, an effective Information Gain Pattern Selection model is designed that minimizes the intrusion detection time and therefore the average delay. Moreover, two new algorithms, namely, an Information Gain Pattern Selection algorithm and Self Adaptive Elliptical Key Generation (SA-EKG) algorithm are proposed for detecting the intruder nodes. The information gain ratio for the patterns is evolved through which the less important patterns are removed. With these patterns as the basis, self adaptive elliptical key generation is applied to detect intruder node and provide reliable data transmission in MANET and therefore improving the packet delivery rate. Simulation results show that the proposed method significantly reduces the average delay and overall packet delivery rate in addition to maintaining a high intrusion detection rate. Top Keywords Mobile Ad-hoc Networks, Intrusion Detection, Information Gain, Self Adaptive, Elliptic Key. Top |