Contents I. Introduction
Markov jump systems are perfect for characterizing dynamic systems with abrupt parameters or structure variations resulting from external causes in the practical working environment. The changes of dynamic systems can be described as the switching behaviors of finite models or subsystems, which are governed by the transition probability matrices in a hidden Markov chain. Because of the excellent ability to model various complex practical systems, applications of Markov jump systems can be found in numerous fields, such as communication, networked control, economic systems, and power systems [1]–[3]. In consequence, Markov jump systems have attracted much attention and a wealth of significant results have been reported in [4]–[8]. Shen et al. [5] has investigated the quantized control of linear Markov jump systems and the disturbance frequency has been incorporated into system synthesis and analysis. The work in [7] is to design an controller with the event-triggered scheme for Takagi–Sugeno (T–S) Markov jump systems, which are subjected to asynchronous phenomena. Furthermore, it is not difficult to synthesize and analyze the design of mode-dependent filter for Markov jump systems. However, practical systems inevitably suffer from some uncertain factors, such as data dropouts and environmental disturbance, which results in the asynchronous relationship between the filter and estimated system. Hence, more and more attention has been transferred to asynchronous filtering design [9]–[11]. Considering the randomly occurred nonlinearity in sensor, Wu et al. [9] have proposed an asynchronous filter for the estimated Markov jump system. With the help of T–S fuzzy techniques, a general design method of asynchronous filter has been presented in [10] for nonlinear Markov jump systems.
