We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Blockchain-enabled federated learning for prevention of power terminals threats in IoT environment using edge zero-trust model.
- Authors
Al Shahrani, Ali M.; Rizwan, Ali; Sánchez-Chero, Manuel; Cornejo, Lilia Lucy Campos; Shabaz, Mohammad
- Abstract
The continuous deepening of information technology in the power industry has dramatically increased the exposure of the power Internet of things. Attackers use the compromised terminal as a springboard to infiltrate the network and can steal sensitive data in the power industry system or carry out damage. Facing the bottleneck of zero-trust centralized deployment of massive power terminal access, an edge zero-trust model is proposed. A cross-context recommendation strategy based on self-attention-based federated learning (SAFL) is proposed in this work. In contrast to current algorithms, SAFL fully mixes the target and auxiliary background information. FL is a technological solution that permits machine learning on-device without transferring the user's private data to a centralized cloud. Therefore, federated learning can help to accomplish personalization. Around the dumb power terminals, the zero-trust engine is deployed in a distributed multi-point, and the trust factors are collected in real-time and stored on the chain. By maintaining a consortium blockchain, the trust factor blockchain (TF_chain), the storage-type edge server synchronizes and shares the trust factor generated by the power terminal during the movement which is convenient for tracing the source and preventing the information from being tampered with abnormal and sensitive elements. Furthermore, lightweight encryption is adopted to ensure authentication information is transmitted from edge to cloud security. The simulation results show that the proposed model can decentralize the zero-trust processing load of centralized deployment and effectively combat the threat of compromised terminals under the condition of marginalized deployment.
- Subjects
FEDERATED learning; BLOCKCHAINS; INTERNET of things; MACHINE learning; CHAIN stores
- Publication
Journal of Supercomputing, 2024, Vol 80, Issue 6, p7849
- ISSN
0920-8542
- Publication type
Article
- DOI
10.1007/s11227-023-05763-6