Improved control of cyber-physical systems subject to cyber and physical attacks

Cyber-Physical Systems (CPS) are defined as integrations of computation, communication, and control in order to achieve the desired performance of physical processes. Security threats have a high possibility of affecting CPS by several cyber attacks without providing any indication about failure. One important problem, especially in power systems, is the control problem of systems under cyber attacks. In this paper, a secure observer-based controller for discrete-time CPS subject to both cyber (denial of service (DoS) and deception) and physical attacks will be presented. The occurrences of the cyber and physical attacks will be considered as Bernoulli distributed white sequences with variable conditional probabilities. A sufficient condition is first derived under which the observer-based controller is guaranteed to have the desired security level using the stochastic analysis techniques. Then, the observer and controller gains will be designed by solving a linear matrix inequality using YALMIP and MATLAB. Finally, an illustrative example is provided to demonstrate the feasibility of the proposed system.