Reinforcing Industry 4.0 With Digital Twins and Blockchain-Assisted Federated Learning
- 30 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (11) , 3504-3516
Abstract
The Internet of Things (IoT) has revolutionized the manufacturing process in the industry. It has created a new ecosystem allowing a diversified set of devices to be controlled remotely with minimal human intervention. Today, with the advances in intelligence, processing, storage, communication, and networking capabilities of IoT devices, we are one step closer to realizing the vision of Industry 4.0. Cyber-physical systems (CPS) are now significantly more intelligent and automated with the aid of advances in Machine Learning (ML). Intelligent IoT (IIoT), Digital Twins (DT) and the advances in mobile networks are now paving the path towards decentralized self-managed CPS in the industry. DT permits mobile networks to provide adaptive and dynamic configurations for cooperative CPS. Moreover, trustworthy cooperation may be realized with blockchain. In this article, we present a blockchain-assisted hierarchical federated learning (FL)-enabled platform (HFL) for Industry 4.0. The solution integrates DT into CPS to accurately capture the characteristics of industrial IoT devices and assist in the HFL process. A two-stage FL algorithm is used that groups Internet-enabled factory machinery and their DTs into groups in accordance with their organizational structure. A global model is created for the groups from the averaged local models and the DT model in the first stage. During the second stage, federated aggregation is used to create a global model from the first-stage models. Blockchain is used to cross-verify and validate newly added blocks with the support of validator nodes. Numerical analysis is performed to compare between the presented DT-enabled and blockchain-assisted HFL solution and benchmark solutions in terms of network overhead, block optimization, and accuracy.Keywords
This publication has 34 references indexed in Scilit:
- Blockchain and Fog Computing for Cyberphysical Systems: The Case of Smart IndustryComputer, 2020
- A Blockchained Federated Learning Framework for Cognitive Computing in Industry 4.0 NetworksIEEE Transactions on Industrial Informatics, 2020
- A Hierarchical Blockchain-Enabled Federated Learning Algorithm for Knowledge Sharing in Internet of VehiclesIEEE Transactions on Intelligent Transportation Systems, 2020
- Client-Edge-Cloud Hierarchical Federated Learning2020
- Providing secure and reliable communication for next generation networks in smart citiesSustainable Cities and Society, 2020
- Blockchain-Assisted Secure Device Authentication for Cross-Domain Industrial IoTIEEE Journal on Selected Areas in Communications, 2020
- A Trustworthy Privacy Preserving Framework for Machine Learning in Industrial IoT SystemsIEEE Transactions on Industrial Informatics, 2020
- Energy-Efficient and Coverage-Guaranteed Unequal-Sized Clustering for Wireless Sensor NetworksIEEE Access, 2019
- A Distributed Synchronous SGD Algorithm with Global Top-k Sparsification for Low Bandwidth Networks2019
- An intrusion detection system for connected vehicles in smart citiesAd Hoc Networks, 2019