A TRUST-AWARE FEDERATED MULTI-MODEL MACHINE LEARNING FRAMEWORK FOR INTELLIGENT DDOS ATTACK DETECTION AND NETWORK SECURITY ENHANCEMENT

Abstract

The rapid growth of Internet of Things (IoT) devices has significantly increased the vulnerability of modern networks to sophisticated cyber threats, particularly Distributed Denial-of-Service (DDoS), Denial-of-Service (DoS), Backdoor, Injection, and Reconnaissance attacks. Existing intrusion detection systems often suffer from limited scalability, inadequate adaptability, and poor trust management in distributed IoT environments. To address these challenges, this study proposes a Hybrid Attention-Driven Federated Framework integrating machine learning, deep learning, federated learning, and blockchain-assisted trust verification for intelligent IoT cyber threat detection. Three advanced models, namely AERF-XGBNet, SHADE-Net, and BAFID, are developed using attention mechanisms, ensemble intelligence, self-healing learning, and federated consensus strategies. Experimental evaluation on the Bot-IoT and ToN-IoT datasets demonstrates superior detection performance, achieving over 99% accuracy, precision, recall, and F1-score while maintaining robust scalability and reliability.