Hidden Markov model‐based filtering for 2‐D Markov jump Roesser systems subject to analog fading channels

Abstract

This article considers the filtering problem for 2‐D Markov jump Roesser systems, in which the measurement signal is transmitted via analog fading channels. The time‐varying amplitude attenuation of the analog fading channels is described by the fading gain modeled as a random variable. The hidden Markov model is used to deal with the mode mismatch phenomenon between the plant and the filter, then a full‐order asynchronous filter is designed for 2‐D Markov jump Roesser systems under analog fading channels. The purpose is to introduce a filter to guarantee that the filtering error system is not only asymptotically mean square stable, but also satisfies the disturbance attenuation performance. An improved decoupling method is proposed to acquire the 2‐D filter parameters such that the conservatism of the 2‐D filter designed law can be effectively reduced. Finally, a verification example and Darboux equation are given to illustrate the validity of the introduced asynchronous filter design law.