Detection and Mitigation of Distributed Denial of Service (DDoS) Attacks on Software Defined Networks (SDNs) Using Multilayer Perceptron (MLP)
Rian Fahrizal, Ivan Munandar, Fadil Muhammad · 2024
Distributed Denial of Service (DDoS) attacks on Software Defined Networks (SDNs) have become a growing threat, resulting in serious disruptions to service availability and infrastructure reliability. Currently, the most common DDoS attack detection technique used on both traditional and SDN networks is the statistic-based detection method, but this method has a weakness in detecting new or unprecedented attacks. This can be overcome by using deep learning methods that are able to handle nonlinear, heterogeneous, and high-dimensional data, which are often found in network traffic data. This research develops a DDoS attack detection and mitigation system on SDN network traffic using the Multilayer Perceptron (MLP) algorithm. The tests conducted show that the DDoS attack detection and mitigation system developed in this study has an accuracy level of 71.17%. This system has proven to be effective in detecting DDoS attacks. Post-mitigation network performance analysis shows a significant improvement in performance compared to pre-mitigation conditions. Previously, bottlenecks were identified in the tissues. However, after mitigation, the network can return to optimal functioning.