Anomaly Detection in IPv4 and IPv6 networks using machine learning

Abstract

Anomaly Detection is an important requirement to secure a network against the attackers. Detecting attacks within a network by analysing the behaviour pattern has been a significant field of study for several researchers and application systems in IPv4 as well as IPv6 networks. For precise anomaly detection, it is essential to implement and use an efficient data-mining methodology like machine learning. In this paper, we contemplated an anomaly detection model which uses machine learning algorithms for data mining within a network to detect anomalies present at any time. This proposed model is evaluated against Denial of Service (DOS) attacks in both IPv4 and IPv6 networks while selecting the most common and evident features of IPv6 and IPv4 networks for optimizing the detection. The results also show that the proposed system can detect most of the IPv4 and IPv6 attacks in efficient manner.