Analysis of Replay Attacks With Countermeasure for State Estimation of Cyber-Physical Systems

This brief studies the security issue of remote state estimation of cyber-physical systems under replay attacks, where the measurement residual is computed in smart sensors and then transmitted to a remote estimator. Firstly, the compromised estimation error covariance is derived to analyze the effect of replay attacks on system estimation performance. Next, a state estimate based coding detection scheme is proposed, which utilizes the estimate information to encode and decode the measurement residual. Compared with existing works, the developed coding scheme only uses the system estimate information, not introduces extra coding information. Furthermore, the design principle of the coding matrix is deduced for the sake of being able to assist a residual-based detector to successfully expose such attacks. Finally, simulation results are provided to verify the effect of replay attacks on the system performance and the effectiveness of the developed coding scheme.