Title: Support vector machine and fuzzy-based intrusion detection and prevention for attacks in MANETs

Authors: A. Anna Lakshmi; K.R. Valluvan

Addresses: Department of CSE, KSR College of Engineering, Tiruchengode, Tamil Nadu, India ' Department of ECE, Velalar College of Engineering and Technology, Erode, Tamil Nadu, India

Abstract: In mobile ad hoc networks, the isolation of critical routing nodes may affect the routes and re-routing that result in significant routing disruption. This in turn degrades the network performance considerably. The existing anomaly-based detection and prevention technique does not involve any standard training technique like neural networks, support vector machines (SVMs) and so on. In order to overcome these issues, in this paper, it is proposed to design a SVM and fuzzy-based intrusion detection and prevention for MANET attacks. Initially, the nodes with the maximum stability index are chosen as cluster heads (CH) and the other nodes become cluster members (CM). The secure communication is established between CH and CMs. Then a support vector machine is utilised to distinguish misbehaving nodes from well-behaved nodes. Then fuzzy rules are used to isolation of misbehaving nodes to prevent intrusion. By simulation result, it is shown that the proposed technique enhances the network performance.

Keywords: mobile ad hoc networks; MANETs; intrusion detection; support vector machines; SVM; cluster heads; cluster members; maximum stability index; fuzzy logic; attack prevention; secure communication; network security; simulation.

DOI: 10.1504/IJMNDI.2015.072837

International Journal of Mobile Network Design and Innovation, 2015 Vol.6 No.2, pp.63 - 72

Received: 15 Mar 2014
Accepted: 20 Sep 2014

Published online: 04 Nov 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article