An Efficient DDoS Attacks Detection And Mitigation Approach In SDN-IoT Network

The Internet of Things (IoT) has revolutionized modern life by introducing advanced solutions and enhancing processing power. However, managing IoT networks can be challenging due to the multitude of devices and communication protocols involved. Software-Defined Networking (SDN) offers a potential solution by simplifying network management, providing network abstraction, facilitating network development, and potentially addressing the complexities of IoT networks. Despite the numerous benefits, it is important to note that IoT systems still face significant security concerns. This study uses entropy as a metric on the SDN controller in an IoT network to assess the homogeneity of network traffic. And on the SDN controller network log data is present so to filter network traffic, the technique additionally incorporates machine learning (ML) classifiers. We tested the suggested methodology’s effectiveness using real-world network traffic information. SDN-IoT networks and ML are used to improve the process of detection and mitigation. The proposed method has a detection rate between 98% and 100%, demonstrating its exceptional performance. Compared to existing methods, our approach not only detects Distributed Denial of Service (DDoS) attacks with remarkable accuracy, but also enables early detection while maintaining a remarkable false-positive rate. This combination of high precision and early detection enables us to effectively identify and mitigate DDoS attacks, thereby enhancing security and minimizing the potential harm caused by malicious activities.