Intelligent DDoS Detection System in Software-Defined Networking (SDN)

Software defined network (SDN) is a network paradigm that decouples the control and data plane of network layer by introducing a logically centralized control known as controller. SDN promises network monitoring, easy and instant network implementation and effective central control but at the same time this central control introduces a single point of failure thereby increasing the risk of Distributed Denial of Service (DDoS) attacks. Thus, this paper analyses the various types of DDoS attacks that can be performed in SDN environment with the help of various attack models. Further, an attempt is made to analyze such attacks and bring out the features that could uniquely classify legitimate traffic and attack traffic using machine learning techniques. Finally, an Open Network Operating System (ONOS) controller module is developed that could live detect any ongoing DDoS attack.