Machine Learning-Based Approach for Detecting DDoS Attacks in Software Defined Networks
Abeer Hakeem, Afraa Attiah · International Journal of Computer Applications · 2024
Software-Defined Networking (SDN) provides enhanced manageability, control, and dynamic updating of network rules through the separation of the control and data planes.However, SDN architectures remain vulnerable to various network attacks, including Distributed Denial of Service (DDoS) attacks.To address this challenge, this paper proposes the DDoSDetect solution, which leverages Logistic Regression machine learning algorithm to detect DDoS attacks in SDN environments.The DDoSDetect solution focuses on identifying flooding-based DDoS attacks, including TCP SYN, HTTP, UDP, and ICMP attacks, by analyzing SDN network traffic.The Logistic Regression classifier is trained to distinguish between normal and attack traffic based on four key features: number of packets, packet size, source and destination MAC addresses.The performance of the DDoSDetect solution is evaluated and compared to other binary classification algorithms, such as Naive Bayes, Random Forest, K-Nearest Neighbor and Support Vector Machine.The experimental results demonstrate that the DDoSDetect solution based on logistic regression outperforms the well-known performing alternative classifiers, achieving an accuracy improvement of 2.4%, an F1-score enhancement of 2.0%, and a precision increase of 11.68%.