XGBoost Machine Learning Model-Based DDoS Attack Detection and Mitigation in an SDN Environment
Tanvi Arvind, K A Radhika · International Journal of Engineering Trends and Technology · 2023
SDN sparked tremendous interest because of its several benefits, such as simple programming, quick scalability, centralized administration, etc. However, security is a significant problem, and Distributed denial of service (DDoS) threats a major challenge for SDN. One way to safeguard a Software-Defined networking infrastructure from DDoS assaults is to use machine learning models. This study presents an XGBoost-based approach for DDoS detection and mitigation. It evaluates it against other Machine Learning techniques, including Logistic Regression, Naive Bayes, Decision Trees, XGBoost, and Multilayer Perceptron. This method will generate, collect, classify, detect, and then mitigate Distributed denial-of-service assaults. The results show that the suggested approach protects SDN from DDoS attacks with high accuracy and a low error level while making good use of network resources. Despite the short training and testing period, the proposed method detects DDoS attacks with greater accuracy.