DDoS Detection in SDN using Machine Learning Techniques

Software-defined network (SDN) becomes a new revolutionary paradigm in networks because it provides more control and operation over infrastructure. The SDN controller is considered as the operating system of based infrastructure, responsible for executing different applications maintaining services functionalities. Despite all its tremendous capabilities, face many security issues due to complexity architecture. Distributed denial (DDoS) common attack on centralized architecture, especially at layer that has network-wide impact. Machine learning now widely used fast detection these attacks. In this paper, some important feature selection methods machine DDoS are evaluated. optimal features reflects classification accuracy techniques performance controller. A comparative analysis classifiers also derived detect experimental results show Random forest (RF) classifier trains accurate model with 99.97% using subset by Recursive elimination (RFE) method.