New approach for Intrusion Detection in Mobile Ad-hoc Networks

As Mobile Ad-hoc network (MANET) has become a very important technology, research concerning itssecurity problem, especially, in intrusion detection has attracted many researchers. Feature selectionmethodology plays a central role in the data analysis process. This article proposes features that effectivelypresent network’s operating condition. The proposed approach is aimed to address the problem ofredundant and irrelevant features that often reduce performance of the detection system. Performancereduction will occur both in speed and predictive accuracy. This paper aims to select and analyze thenetwork features using principal component analysis.Performing various experiments, normal and attack states are simulated and the results for the selectedfeatures are analyzed and presented. Classification and clustering methods are used to detect networkstatus. In this way, paper intends to propose a method of approach which can generate an acceptable resultwith high accuracy for all types of the Ad-hoc networks.

Keywords:

Ad-hoc , feature selection , classification , clustering , Intrusion detection