DDoS Attack Detection Using Machine Learning Techniques in SDN

Software Defined Networks (SDN) paradigm was introduced to overcome the limitations of the traditional network. It becomes a promising network architecture that provide network operators more control over the network infrastructure. The controller also called as the operating system of the SDN which has the centralized control over the network. Despite all its capabilities, introduction of various architectural entities of SDN poses many security threats. Among many such security threats, Distributed Denial of Services (DDoS) is a rapidly growing attack. This targets the availability of the network, by flooding the controller with spoofed packets. Therefore, it is important to design a robust attack detection mechanism to prevent the control plane DDoS attack. In this work, we have used Machine Learning techniques such as Naive Bayes, Random Forest, Multilayer Perceptron and Support Vector Machines to classify and predict DDoS attacks like ICMP-Echo, Smurf, TCP SYN, and HTTP flood on a self generated dataset. Experimental results with proper analysis have been presented in this work.