Blockchain-based zero trust networks with federated transfer learning for IoT security in industry 5.0.

Abstract

The rise of Industry 5.0 focuses on merging advanced intelligence, automation, and human-centered teamwork in industrial settings. However, keeping interconnected IoT networks secure is still a challenging problem. This paper proposes a new security framework that combines Blockchain, Federated Transfer Learning, and zero trust network (ZTN) principles to improve IoT security in Industry 5.0. Blockchain is a decentralized ledger that ensures secure data sharing and protects model updates. Federated Transfer Learning allows model training across distributed IoT devices to keep data private. The ZTN approach enforces strict access rules, assuming that no entity is trusted by default. The proposed framework offers a scalable and resilient solution to protect next-generation industrial IoT networks, using Blockchain for data security, transfer learning for adaptability, and ZTN for strict access control. The ZTN architecture strengthens security by checking every access request and keeping the IoT system safe. The experimental results show good performance of the proposed method, with better accuracy, precision, recall, and F1 scores. The model achieved an accuracy of 0.85, 0.88, and 0.87 for learning rates of 0.01, 0.001, and 0.0001, respectively, at 100 epochs. The precision values reached 0.84, 0.87, and 0.86, while the recall scores were 0.82, 0.86, and 0.85, respectively. The F1-scores were recorded at 0.83, 0.86, and 0.85, which confirms the robustness of our model.