Detection of DoS Attack Using Machine Learning in Software Defined Network

Detection of DoS Attack Using Machine Learning in Software Defined Network

연구 분야: Networking

학회: International Conference on Advances in Artificial Intelligence and Machine Learning in Big Data Processinging

Nowadays, Software-Defined Networking (SDN) offers benefits in the areas of automation, elasticity, and resource consumption. However, evidence is there that SDN controllers may undergo certain defeat for the network structure, particularly as they are targeted by attacks like Denial of Service (DoS) attacks. Due to this network traffic has increased tremendously and attacked the server severely. To handle this issue, we used the RYU controller and Mininet tool to identify and alleviate the DoS attack by the Machine learning algorithm. Since machine learning (ML) is deemed as the main method for detecting peculiarities, the detection of DoS attacks was done through Machine learning-based classification. In this paper, several machine learning techniques were used to identify the DoS attack, and the traffic which is causing the attack has been dropped immediately to avoid congestion. The proposed work has been simulated in Mininet and the results show that the proposed work detects DoS attacks well and achieves good accuracy.