Journal of Innovation in Computer Science and Engineering
Year : 2012, Volume : 2, Issue : 1
First page : ( 29) Last page : ( 33)
Print ISSN : 2278-0947. Online ISSN : 2455-3506.

Black Hole Detection in Adhoc Networks using Neural Networks

Dharmar Vydeki*, Parveen M. A. Ayisha Razeena**, Bhuvaneswaran R.S.***

Eswari Engineering College, Anna University, Chennai

*dvydeki@rediffmail.com

**mujarazee@gmail.com

***bhuvan@annauniv.edu

Online published on 27 June, 2017.

Abstract

Intrusion Detection System (IDS) plays an important role in providing additional security to the challenging Mobile Ad hoc networks (MANETs). Using computational intelligent techniques to the detection process is proved to be more suitable as they have human-like decision-making capabilities. This paper proposes a combinatorial approach to intrusion detection for MANETs to detect black hole attack, by combining anomaly and specification-approaches of IDS. The proposed system aims at designing IDS using one of the fundamental soft computing mechanisms: artificial neural network (ANN). The design of IDS is tested with simulated MANETs for various traffic conditions. The performance of the system is analyzed based on the true and false positive rates. It is deduced from the experimental results that the NN-based system produces 100% true positive rate.

Top

Keywords

AODV, Black Hole Attack, Anomaly IDS, Specification based IDS, ANN.

Top