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

Area of Science:

• Industrial IoT Security
• Cybersecurity Frameworks
• Industry 5.0 Technologies

Background:

• Industry 5.0 emphasizes human-machine collaboration, increasing the complexity and vulnerability of interconnected Industrial Internet of Things (IIoT) networks.
• Securing these IIoT networks against evolving threats remains a significant challenge, hindering the full potential of Industry 5.0.

Purpose of the Study:

• To propose and evaluate a novel security framework for Industry 5.0 IIoT networks.
• To enhance the security, privacy, and resilience of IIoT systems through integrated advanced technologies.
• To address the challenges of data security, privacy, and access control in distributed industrial environments.

Main Methods:

• Integration of Blockchain for secure data sharing and model update protection.
• Implementation of Federated Transfer Learning for privacy-preserving model training across distributed IoT devices.
• Adoption of Zero Trust Network (ZTN) principles for stringent, default-denied access control and continuous verification of all network requests.

Main Results:

• The proposed framework demonstrated robust performance in securing Industry 5.0 IIoT networks.
• Experimental results indicated significant improvements in accuracy (up to 0.88), precision (up to 0.87), recall (up to 0.86), and F1 scores (up to 0.86).
• The framework provides a scalable and resilient solution, confirming its effectiveness through rigorous testing.

Conclusions:

• The combined approach of Blockchain, Federated Transfer Learning, and ZTN offers a powerful solution for enhancing IIoT security in Industry 5.0.
• The framework effectively balances data security, privacy preservation, and adaptive access control.
• The validated performance metrics confirm the robustness and suitability of the proposed security framework for next-generation industrial networks.