Detection of DDoS Attacks in Software Defined Network using Decision Tree

Software Defined Networks (SDN) is a programmable network which can be easily managed with a global knowledge of network topology. However, SDN controller is vulnerable to distributed denial of service (DDoS) due to its centralized architecture. DDoS attacks are dangerous and threatening attacks as they flood the controller with large volume of packets. This leads to the failure of SDN controller which is a critical issue of security. In this work, we initially detect the different types of DDoS attacks using classification algorithms for CIC-DDoS 2019 dataset. Next, we capture packets from SDN environment, pre-process the data and apply classification algorithm to detect DDoS attacks using SDN dataset. We create SDN dataset with Mininet emulator and RYU controller using different DDoS tools. The results reveal that decision tree has better performance compared to SVM and Naïve Bayes.