Finite-time secure state estimation for a class of switched systems subject to deception attacks

This article is concerned with the finite-time secure state estimation problem for switched systems subject to malicious attacks. The sensor data is considered transmissible via a non-secure communication network environment, where the communication channel might be subject to deception attacks. By dividing the sensor data that is subject to attacks into two parts, related and unrelated to the original measurements, a switched secure state estimation error system model is established. Moreover, based on the average dwell time method, sufficient conditions are derived for the switched secure state estimation error system to be finite-time stable, as well as finite-time boundedness. An existence condition of the finite-time secure state estimator gains with a desired H infinity$$ {H}_{\infty } $$ performance is given. Subsequently, the finite-time secure state estimator gains' design is transformed to solve a set of linear matrix inequalities. Finally, numerical examples are used in order to show the effectiveness of the proposed finite-time secure state estimator design method.