Design an anomaly-based intrusion detection system using soft computing for mobile ad hoc networks

The main objective of an intrusion detection system is to classify the normal and suspicious activities in the network. The complex characteristics of mobile ad hoc networks make intrusion detection more difficult than for conventional networks. Although, soft computing techniques-based intrusion detection systems proved their effectiveness on wired networks in terms of detecting known and unknown attacks but use of soft computing techniques for mobile ad hoc networks still very restricted so that in this paper, a new scheme has been proposed by using neuro-fuzzy classifier in binary form for mobile ad hoc networks to identify the behaviour of current activities, i.e., normal or abnormal. Qualnet simulator and MATLAB toolbox are used to visualise the attack-based scenarios and evaluate the performance of proposed approach. Simulation results show that the proposed soft computing-based approach is able to identify the known and unknown attacks in mobile ad hoc networks with high positive and low false positive rates.