SDN Based DDOS Detection and Mitigation

Distributed Denial of Service (DDoS) attacks con- tinue to threaten the performance and availability of today’s network infrastructure. Traditional security solutions often fall short in dynamic environments, as they rely on fixed rules and lack the flexibility to respond to evolving threats in real time. This paper introduces a smart and adaptive solution that combines Software Defined Networking (SDN) and Machine Learning (ML) to detect and counter DDoS threats efficiently. Leveraging the RYU SDN controller, Mininet for emulated environments, and a Random Forest classifier for traffic analysis, the system dynamically identifies and mitigates abnormal traffic patterns. Experimental evaluation shows that the framework achieves a detection accuracy of 97.2%, maintains a low false positive rate of 3.1%, and introduces minimal delay, confirming the practicality and effectiveness of the proposed approach.