Asynchronous filtering for Markov jump systems within finite time: A general event-triggered communication

Abstract

In this paper, an event-based asynchronous H∞ filter design method for Markov jump systems over a finite-time interval is proposed. First, the hidden Markov model with a known conditional probability matrix is introduced to describe the asynchronous phenomena due to the failure of precisely obtaining the system mode information. Accordingly, an asynchronous event-triggered transmission strategy with a general threshold is presented to availably save the communication costs. Then, via using the mode-dependent Lyapunov function, a full-order filter regulated by the event-triggered strategy is established to track the system outputs. Based on Finsler’s lemma, sufficient conditions with less conservatism are deduced such that the resulting system is stochastically finite-time bounded while satisfying prescribed H∞ performance. Finally, a simulation example is implemented to illustrate the effectiveness of the theoretical results.