Robust FDI Attack Detection for Cyber-Physical Systems with Noisy and Uncertain Parameters

This paper investigates the problem of False Data Injection attack detection in cyber-physical systems under the influence of noise disturbances and parameter uncertainties. A novel cyber-physical systems model incorporating uncertain parameters is established, with a focus on scenarios where attacks target the system actuators. By leveraging Lyapunov function theory and Linear Matrix Inequality techniques, the gain design conditions for the attack detection estimator are derived, and its stability and robustness are rigorously analyzed. The research results show that even under the influence of noise interference and system parameter uncertainty, the designed attack detection estimator can still accurately identify and detect FDI attacks suffered by the system, thereby effectively improving the system’s security and anti-interference ability. Finally, an example is provided to verify the effectiveness of the method proposed in this paper.