【Author】
Singh, Saurabh; Rathore, Shailendra; Alfarraj, Osama; Tolba, Amr; Yoon, Byungun
【Source】FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
【Abstract】With the dramatically increasing deployment of IoT (Internet-of-Things) and communication, data has always been a major priority to achieve intelligent healthcare in a smart city. For the modern environment, valuable assets are user IoT data. The privacy policy is even the biggest necessity to secure user's data in a deep-rooted fundamental infrastructure of network and advanced applications, including smart healthcare. Federated learning acts as a special machine learning technique for privacy preserving and offers to contextualize data in a smart city. This article proposes Blockchain and Federated Learning-enabled Secure Architecture for Privacy-Preserving in Smart Healthcare, where Blockchain-based IoT cloud platforms are used for security and privacy. Federated Learning technology is adopted for scalable machine learning applications like healthcare. Furthermore, users can obtain a well-trained machine learning model without sending personal data to the cloud. Moreover, it also discussed the applications of federated learning for a distributed secure environment in a smart city. (c) 2021 Published by Elsevier B.V.
【Keywords】Federated Learning; Privacy-preserving; Blockchain; Internet-of-Things
【标题】使用联邦学习和区块链技术保护物联网医疗数据隐私的框架
【摘要】随着物联网（Internet-of-Things）和通信的部署急剧增加，数据一直是在智慧城市中实现智能医疗的重中之重。对于现代环境，有价值的资产是用户物联网数据。隐私政策甚至是在根深蒂固的网络基础设施和高级应用程序（包括智能医疗保健）中保护用户数据的最大必要条件。联邦学习作为一种特殊的机器学习技术来保护隐私，并提供在智慧城市中对数据进行情境化。本文提出了区块链和联邦学习支持的安全架构，用于智能医疗中的隐私保护，其中基于区块链的物联网云平台用于安全和隐私。联邦学习技术被用于医疗保健等可扩展的机器学习应用程序。此外，用户无需将个人数据发送到云端即可获得训练有素的机器学习模型。此外，还讨论了联邦学习在智慧城市中分布式安全环境中的应用。 (c) 2021 年 Elsevier B.V. 出版
评论