Security strategy against location-varying sparse attack on distributed state monitoring

Secure state monitoring is mainly concerned with the problem of estimating the state of cyber-physical systems from a set of compromised attacked measurements. Compared with the case of fixed location attacks, a smart attacker who appropriately changes the attack target location can effectively resist intelligent defense detection and enhance the damage effect. To address this erratic attack, this article develops a distributed secure state monitoring protocol for a class of homogeneous cyber-physical networked systems under location-varying sparse attacks. To effectively overcome the influence of variable attack positions on state estimation, a distributed Luenberger-like observer together with a logical detection scheme is proposed such that the remote monitoring center and each local agent can precisely isolate the transmitted data that are being attacked. Sufficient conditions for the existence of observer parameters for secure state monitoring are given in terms of linear matrix inequalities. Finally, a simulation example is provided to verify the proposed theoretical results.