FusionFedBlock: Fusion of blockchain and federated learning to preserve privacy in industry 5.0 br

【Author】 Singh, Sushil Kumar; Yang, Laurence T.; Park, Jong Hyuk

【Source】INFORMATION FUSION

【影响因子】17.564

【Abstract】Nowadays, Industries are experiencing rapid changes in the digital environment, referred to as Industry 5.0. The Internet of Things (IoT) and advanced technologies are essential in the industrial environment. Technological advancements can collect, transfer, and analyze vast amounts of data in the industry via promising technologies. Still, IoT has various issues when applied to industrial infrastructures, such as centralization, privacy preservation, latency, and security. This article proposes a scheme as FusionFedBlock: Fusion of Blockchain and Federated Learning to Preserve Privacy in Industry 5.0 to address the aforementioned issues. At the federated layer, the industry's departments (Production, Quality Control, Distribution) allow local learning updates with network automation and communicate to the global model, which miners verify in the Blockchain networks. Federated-Learning offers privacy preservation between various mentioned departments in industries. Decentralized secure storage is provided by the Distributed Hash Table (DHT) at the cloud layer. The validation outcomes of the proposed scheme demonstrate excellent performance as the accuracy of 93.5% in a 50% active node for Industry 5.0 compared to existing frameworks.

【Keywords】Blockchain; Federated learning; Information fusion; Privacy -preservation; Industrial IoT; Industry 5; 0; Security

【发表时间】2023 FEB

【收录时间】2022-12-18

【文献类型】实证数据

【主题类别】

区块链技术-协同技术-联邦学习