Evaluation of Deep Learning Approaches for Intrusion Detection System in MANET

Abstract

Mobile Ad hoc Network (MANET) consists of a set of nodes which stand randomly in the operating environment. Since nodes are then vulnerable to intrusion and attack without any pre-defined infrastructure and flexibility. Securing in this type of network is an important area. To tackle these security problems, current cryptography schemes cannot completely safeguard MANET in terms of new threats and vulnerabilities. By implementing Deep learning techniques in IDS, MANET will be able to adapt complex environments and allow the system to make decisions on intrusion while continuing to learn about their mobile environment. IDS represent the second line of defense against malevolent behavior to MANET since they monitor network activities in order to detect any malicious attempt performed by Intruders. Recently, more and more researchers applied deep neural networks (DNNs) to solve intrusion detection problems. The two major forms of DNN architectures, Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN), are commonly explored to enhance the performance of intrusion detection system. In this paper, we made a systematic comparison of CNN and RNN on the deep learning-based intrusion detection systems, aiming to give basic guidance for DNN selection in MANET.