Taxonomy of Machine Leaning Based Anomaly Detection and its suitability

Abstract

Mobile Adhoc Networks (MANETs) are gaining popularity due to their potential of providing low cost solutions to real-world communication problems. MANET poses several inherent challenges to intrusion detection due to its distinct characteristics such as its dynamic topology, lack of node management centrally and sustenance in highly resource constrained environment. Intrusion Detection Systems (IDS),analyse/monitor the traffic packets using soft-computing techniques to detect the malicious activities or intrusion for securing MANETs against attackers. Machine learning based soft-computing techniques in IDS are capable of adapting the dynamic environments of MANETs and enables system to make decisions on intrusion while continuing to learn about their mobile environment. In this paper, taxonomy of machine-learning-based-intrusion-detection systems has been presented along with the discussion of its suitability in MANETs. Appropriately deployment and establishment of Machine learning techniques based intrusion model for MANETs are also discussed.