文献类型：期刊文献

英文题名：Security and Privacy-Enhanced Federated Learning for Anomaly Detection in IoT Infrastructures

作者：Cui, Lei[1,2];Qu, Youyang[3];Xie, Gang[1];Zeng, Deze[4];Li, Ruidong[5];Shen, Shigen[6];Yu, Shui[2]

机构：[1]Taiyuan Univ Sci & Technol, Key Lab Adv Control & Intelligent Informat Syst, Taiyuan 030024, Shanxi, Peoples R China;[2]Univ Technol Sydney, Sch Comp Sci, Ultimo, NSW 2007, Australia;[3]Deakin Univ, Sch Informat Technol, Burwood, Vic 3125, Australia;[4]China Univ Geosci, Sch Comp Sci, Wuhan 430074, Hubei, Peoples R China;[5]Kanazawa Univ, Inst Sci & Engn, Fac Elect Informat & Commun Engn, Tokyo 1848795, Japan;[6]Shaoxing Univ, Dept Comp Sci & Engn, Shaoxing 312000, Peoples R China

年份：2022

卷号：18

期号：5

起止页码：3492

外文期刊名：IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS

收录：SCI-EXPANDED(收录号：WOS:000752019100062)、、EI(收录号：20213510842262)、Scopus(收录号：2-s2.0-85113860766)、WOS

基金：This work was supported in part by the Shanxi Province Science Foundation for Youths under Grant 201901D211306, in part by the Australia Australian Research Council (ARC) under Grant DP180102828 and Grant DP200101374, in part by the Central Government Guides Local Science and Technology Development Funds under Grant YDZX20191400002270, and in part by the Key Research and Development Plan Project (International Cooperation) of Shanxi Province under Grant 201803D421039. Paper no. TII21-1732. (Lei Cui and Youyang Qu are co-first authors.)

语种：英文

外文关键词：Anomaly detection; Servers; Internet of Things; Blockchains; Privacy; Collaborative work; Security; Asynchronous federated learning; differential privacy protection; IoT anomaly detection; security

外文摘要：Internet of Things (IoT) anomaly detection is significant due to its fundamental roles of securing modern critical infrastructures, such as falsified data injection detection and transmission line faults diagnostic in smart grids. Researchers have proposed various detection methods fostered by machine learning (ML) techniques. Federated learning (FL), as a promising distributed ML paradigm, has been employed recently to improve detection performance due to its advantages of privacy-preserving and lower latency. However, existing FL-based methods still suffer from efficiency, robustness, and security challenges. To address these problems, in this article, we initially introduce a blockchain-empowered decentralized and asynchronous FL framework for anomaly detection in IoT systems, which ensures data integrity and prevents single-point failure while improving the efficiency. Further, we design an improved differentially private FL based on generative adversarial nets, aiming to optimize data utility throughout the training process. To the best of our knowledge, it is the first system to employ a decentralized FL approach with privacy-preserving for IoT anomaly detection. Simulation results on the real-world dataset demonstrate the superior performance from aspects of robustness, accuracy, and fast convergence while maintaining high level of privacy and security protection.