Mitigation of distributed denial of service (DDoS) attacks over software defined networks (SDN) using machine learning and deep learning techniques

Abstract

Software Defined Networking (SDN) is an emerging networking paradigm which enables network control to be confined to a logically centralized controller. This enables global visibility of network and easier network management. The capability to program network through high level programming languages makes SDN a suitable network model to be extensively deployed in live environments. Still SDN is subject to several network attacks, among which DDoS-Distributed Denial of Service attack is the most prominent one. Controller which is the brain of SDN can be paralyzed by a high scale DDoS attack. Security of SDN is in immature state and considerable research is done in this area by both industry and academia. This paper focuses on the SDN DDoS mitigation techniques using Machine Learning (ML) and Deep Learning (DL) techniques. Network traffic features for determining DDoS are also surveyed in this work.