International Journal of Scientific Research and Engineering Development

Distributed Denial-of-Service (DDoS) attacks represent one of the most pervasive and destructive threats to modern networked infrastructure, capable of rendering services unavailable by saturating bandwidth, exhausting server resources, or disrupting critical routing mechanisms. Traditional, static network architectures are ill-equipped to respond adaptively to such attacks due to their decentralised control and limited visibility. Software-Defined Networking (SDN) addresses these shortcomings by decoupling the control plane from the data plane, enabling centralised, programmable, and real-time network management. This paper proposes a comprehensive adaptive DDoS mitigation framework built on SDN, which integrates continuous traffic monitoring, multi-stage anomaly detection, and dynamic mitigation enforcement. The framework leverages machine learning classifiers—including Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbours (KNN), and Naive Bayes (NB)—alongside ensemble voting mechanisms and statistical analysis to achieve high detection accuracy with minimal false positives. Upon detecting anomalous flows, the SDN controller automatically deploys mitigation actions such as IP blocking, rate limiting, and traffic redirection via OpenFlow-enabled switches. Experimental evaluations on benchmark datasets (CICIDS2017, CICIDS2018, CICDDoS2019) confirm detection accuracies exceeding 98%, while the proposed framework demonstrates superior network resilience, faster incident response, and sustained service availability compared to conventional mitigation approaches.

Vinayak Ghaniger, Aman Pawaskar, Prof.Pavan Mitragotri,"Behaviour-Aware Adaptive DDoS Detection and Mitigation in Software-Defined Network Architecture" International Journal of Scientific Research and Engineering Development, V9(3): Page(958-969) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.