Abstract This research proposes an efficient, lightweight clustering-based traffic classification model for the detection of Distributed Denial of Service (DDoS) attacks in Software-Defined Networking (SDN) environments. The model synergistically integrates K-Means clustering with dimensionality reduction techniques such as Principal Component Analysis (PCA) to improve detection accuracy an‎d computational efficiency. By addressing the limitations of signature-based detection an‎d the scalability issues in current machine learning approaches, this study demonstrates that unsupervised clustering can effectively identify abnormal traffic behaviors indicative of DDoS attacks, including zero-day threats. Dimensionality reduction reduces the feature space complexity, mitigates the curse of dimensionality, an‎d enhances clustering performance, making this model suitable for real-time SDN applications. eva‎luation metrics such as accuracy, detection rate, false positive rate, an‎d clustering validity measures confirm the model’s effectiveness. This research contributes to the development of scalable an‎d adaptable solutions for enhancing the security of next-generation programmable networks.

6/10/2025 12:00:00 AM

86894

1404/04/23

(An Efficient Clustering-Based Traffic Classification Model for DDoS Attack Detection in SDN Using K- Means an‎d Dimensionality Reduction)

Software-Defined Networking (SDN) , Distributed Denial of Service (DDoS) , K-Means Clustering, Dimensionality Reduction , Principal Component Analysis (PCA) , Traffic Classification , Anomaly Detection , Network Security , Machine Learning , Zero-Day Attack Detection