A Stealthy False Data Injection Attack Scheme Against Sensor Measurements Using Partial System Knowledge

• 成果类型: Article
• 署名作者: Guo, Haibin; Pang, Zhong-Hua; Sun, Jian; Han, Qing-Long; Liu, Guo-Ping
• 署名单位: North China University of Technology; Beijing University of Technology; Beijing Institute of Technology; Swinburne University of Technology; Southern University of Science & Technology
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2025.3550056
• 发表日期: 2025
• 页码: 5600-5607
• 关键词: estimation; Kalman filters; detectors; steady-state; estimation error; ELECTRONIC MAIL; automation; remote control; Cyberattack; actuators; Attack stealthiness; Kalman filter; networked control systems (NCSs); stealthy false data injection (FDI) attacks

摘要

Attack design is indispensable for analyzing potential risks of networked control systems (NCSs). However, the remote control center is usually well protected and its knowledge is difficult to be disclosed, which becomes a major obstacle in developing stealthy false data injection (FDI) attack scheme because only partial system knowledge (i.e., the system matrices of the physical plant) could be used. To meet this challenge, a novel stealthy FDI attack scheme against the sensor measurement is proposed by employing the normal and compromised self-governed filters held by malicious attackers, where the normal one is adopted to estimate the system state and the compromised one is used as the virtual attacked target. The corresponding attack strategy is obtained by maximizing the estimation error of the compromised self-governed filter. Then, the residual of the compromised system is derived to prove attack stealthiness. Next, it is derived and found that the attack impact on system estimation performance is the same as that based on full system knowledge. Furthermore, the divergence condition of NCSs under the attack is presented. Finally, all the theoretical analyses are verified by simulation results.