Attack Detection and Secure State Estimation of Collectively Observable Cyber-Physical Systems Under False Data Injection Attacks

• 成果类型: Article
• 署名作者: Suo, Yuhan; Chai, Runqi; Chai, Senchun; Farhan, Ishrak M. D.; Xia, Yuanqing; Liu, Guo-Ping
• 署名单位: Beijing Institute of Technology; Southern University of Science & Technology
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3316160
• 发表日期: 2024
• 页码: 2067-2074
• 关键词: Attack detection; collectively observable systems; cyber-physical systems (CPSs) security; secure state estimation; Sensor fusion

摘要

In this technical note, the problem of attack detection and secure state estimation in collectively observable cyber-physical systems is considered. First, an attack signal estimator is designed, which theoretically realizes the unbiased estimation of attack signals. Then, the alert, whether the sensor is attacked, is described as a hypothesis testing problem from the perspective of average malicious disturbance power, and a novel attack detection algorithm is designed on this basis. Based on the objective of minimizing the fusion error of each fusion center at each time, an efficient sensor fusion algorithm is proposed. The problem of solving the optimal fusion coefficient matrix is transformed into a linear programming problem, which is solved by the method of Lagrange multipliers. The theoretical results show that the proposed algorithm significantly improves the computational efficiency without compromising the estimation performance. Finally, an example of vehicle target state estimation is given to illustrate the effect of the proposed method.