Privacy protection federated learning system based on blockchain and edge computing in mobile crowdsourcing

Wang, Weilong; Wang, Yingjie; Huang, Yan; Mu, Chunxiao; Tang, Zhiru; Tong, Xiangrong; Cai, Zhipeng

doi:10.1016/j.comnet.2022.109206

Cited by 17 publications

(1 citation statement)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A comprehensive study of blockchain applications in IoT was presented in [13]. author [14] talked about a study of combination of blockchain IoT (BCoT). A rigorous analysis of security of IoT-based blockchain strategies was offered by work [15].…”

Section: Related Workmentioning

confidence: 99%

Soft computing in secure social internet of things based on post-quantum blockchain with federated learning

Bishnoi

Parikh

Kumari

et al. 2023

Preprint

View full text Add to dashboard Cite

A system's compliance with a specified security model, security standard, or specification is the focus of a security evaluation. The process of selecting the appropriate model for security assessment is determined by the type of cryptosystem. There are various models available. By enhancing data authentication and security for soft computing, this study proposes a novel technique for secure social internet of things (SSIoT) privacy analytics using post-quantum blockchain federated learning with encryption and trust analysis, the privacy analysis and data authentication. In terms of latency, QoS, energy consumption, packet loss rate, and other parameters, the experimental analysis is carried out. The purpose of the security analysis and performance evaluations is to demonstrate that the proposed scheme can satisfy the security requirements and enhance the FL model's performance. Federated learning is able to carry out effective machine learning (ML) with multiple participants while maintaining privacy of terminal personal data. Our proposed combination of federated learning as well as blockchain provides a solid foundation for future industrial Internet, as demonstrated by the numerical results.the proposed technique attainedenergy consumption of 55%, packet loss rate of 59%, QoS of 79%, Latency of 72%, network security analysis of 82%.

show abstract

Section: Related Workmentioning

confidence: 99%