Machine Learning Techniques to Detect DDoS Attacks in IoT's, SDN's: A Comprehensive Overview

Attacks known as distributed denial of service (DDoS) compromise user privacy while disrupting internet services and posing a serious danger to network security. DDoS attack detection using machine learning (ML) techniques has showed promise, but the evolving nature of these attacks presents challenges in accurately distinguishing between attack patterns and normal traffic. This paper presents a comprehensive overview of effective ML techniques for DDoS attack detection, focusing on IoTs, SDNs, and cloud. The literature survey analyzes research findings, categorized according to a suggested taxonomy, providing insights into the strengths and limitations of different approaches. Deploying and evaluating ML-based models in real-world environments is crucial to assessing practical effectiveness. This paper highlights the potential of ML techniques in detecting DDoS attacks while emphasizing the need for further research to address evolving attack tactics, establish evaluation practices, and develop adaptive defenses for real-world scenarios. By pursuing these avenues, network systems can significantly enhance security and resilience against DDoS attacks.