An enhanced federated machine learning for secure DDOS detection in IOT network

The rapid growth of Internet of Things (IoT) devices has created new opportunities for automation and connectivity, but it has also increased exposure to cyber-attacks especially Distributed Denial of Service (DDoS) attacks. Traditional centralized security systems struggle to protect IoT networks because they require collecting large amounts of data in one place, which raises privacy concerns and slows down detection. This research proposes an enhanced Federated Machine Learning (FML) approach for secure and efficient DDoS detection in IoT environments. Instead of sending raw data to a central server, each IoT device trains a local model and only shares learned model updates, keeping sensitive information private. The enhanced system combines lightweight learning algorithms, secure communication techniques, and improved aggregation methods to boost accuracy and resistance against malicious interference. Experimental results show that the proposed approach achieves higher detection accuracy and lower latency compared to traditional centralized methods, reduces data exposure, and perform well even in resource-limited IoT devices. Overall, the enhanced FML-based solution provides a stronger, more privacy-preserving, and scalable defence mechanism for securing modern IoT networks against DDoS threats. This study addresses a critical gap in existing research, where many federated learning (FL)-based intrusion detection systems for IoT environments fail to adequately handle data heterogeneity, communication overhead, and robustness against sophisticated distributed denial-of-service (DDoS) attacks. To overcome these limitations, the proposed approach introduces an enhanced federated machine learning framework that integrates adaptive model aggregation and lightweight anomaly detection mechanisms to improve detection accuracy while preserving data privacy. The main contributions of this study include the development of an optimized FL-based detection model, improved resilience against diverse DDoS attack patterns, and reduced communication costs suitable for resource-constrained IoT devices. The remainder of the paper is structured as follows: Section 2 reviews related literature, Section 3 presents the proposed methodology, Section 4 discusses experimental results, and Section 5 concludes the study with recommendations for future research.