Machine Learning Techniques for Detecting DDoS Attacks in SDN
M. S. Kavitha, M. Suganthy, Aniket Biswas, R. Srinivsan, R. Kavitha, A. Rathesh · 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS) · 2022
Future internet is increasingly reliant on Software Defined Networking (SDN). With SDN, networks can be dynamically controlled, providing a global network. Compared to traditional networks, SDN offers the advantage of better security provisioning due to centralized management. However, SDN architecture manifests several new network security problems that need to be handled to improve the security of SDN networks. Information security and data analysis systems for Big Data have become more essential due to the increasing volume of data and its incremental growth. Monitoring and analyzing data is needed to detect any intrusion into a system or network via an intrusion detection system (IDS). By using traditional methods, traditional data analysis techniques are unable to detect attacks caused by high volumes, a wide variety and high speeds of network data. For an accurate and efficient data analysis process, IDS employs Big Data techniques. The paper uses machine learning models to detect Distributed Denial of Service (DDoS) attacks. The machine learning model is trained using data from KDD Cup 99.K Nearest Neighbor Classifier, Logistic Regression, and Decision Tree have been used to train and test the datasets. It can be concluded that machine learning methods can be more effective at detecting DDoS attacks than traditional methods, that can be applied to software defined networks. Several experiments demonstrate the potential of our proposal to detect intrusion in SDN environments after extensive evaluation.