A Comprehensive Review of DDoS Attack Detection in IoT using Deep Learning and Intelligence Techniques

The rapid growth of the Internet of Things (IoT) has expanded connectivity and automation but also increased vulnerability to cyberattacks, especially Distributed Denial of Service (DDoS) attacks. Detecting such attacks in IoT environments is challenging due to large data volumes, device heterogeneity, and limited resources. Recently, machine learning and deep learning techniques have been widely adopted to improve intrusion detection performance. This paper reviews state-of-the-art IoT DDoS detection methods, classifying them into hybrid deep learning models, federated learning frameworks, optimization and feature selection techniques, machine learning and ensemble methods, SDN/edge-based architectures, and blockchain-enabled solutions. These approaches are evaluated based on architecture, datasets, performance metrics, advantages, and limitations. The review finds that hybrid deep learning models achieve high detection accuracy, while federated learning and blockchain approaches enhance privacy and security. Machine learning models are lightweight and efficient, especially when combined with optimized feature selection. However, challenges remain, including high computational complexity, poor generalization, limited explainability, and difficulty detecting zero-day attacks. Finally, the paper identifies research gaps and highlights the need for lightweight, scalable, and adaptive intrusion detection systems suitable for dynamic IoT environments.