Finite-time distributed resilient state estimation subject to hybrid cyber-attacks: a new dynamic event-triggered case

This paper is concerned with the issues of finite-time distributed resilient state estimation subject to hybrid cyber-attacks. The information exchanges among estimators are governed by an improved dynamic event-triggered mechanism, in which the time-varying threshold with predetermined upper and lower bounds is updated by artificial internal dynamics. With the help of the Lyapunov stability theory combined with the S-procedure, a sufficient condition is developed such that the augmented error dynamics are stochastic finite-time bounded. Furthermore, the desired estimator gains are acquired in terms of the solution to certain matrix inequalities which involve the information of communication topology, cyber-attack probabilities as well as the uncertainty of gain matrices. Finally, the effectiveness of the designed distributed state estimator is illustrated by a numerical example.

Automated summary

This paper focuses on finite-time distributed resilient state estimation subject to hybrid cyber-attacks. An improved dynamic event-triggered mechanism is proposed for information exchanges among estimators, with a time-varying threshold updated by artificial internal dynamics. By using Lyapunov stability theory combined with the S-procedure, a sufficient condition is developed for stochastic finite-time boundedness of augmented error dynamics. The desired estimator gains are obtained through the solution to matrix inequalities involving communication topology, cyber-attack probabilities, and uncertainty of gain matrices. The effectiveness of the designed distributed state estimator is illustrated with a numerical example.