Comparison of Various ML Approaches for Detection of DDoS Attacks in SDN

Abstract

The network has become increasingly complicated and diversified in recent years due to the rapid growth of internet technology and its coverage. Software-defined networking (SDN) is a recent approach to network architecture. DDoS attacks have a measurable effect on SDN because of their centralized nature. This results in an early discovery, and prevention of DDoS attacks is essential. DDoS is a network risk that tries to flood targeted networks with unwanted data. This paper focuses on various ML approaches for classifying DDoS attacks by providing decision-making. That results in making the computing system more intelligent. We have used the most popular ML techniques to identify the attacks. To prevent and mitigate attacks, the evaluation process includes training and adoption of a suitable model for the specific network.