Asynchronous filtering for discrete-time T–S fuzzy sampled-data systems with improved quantization

Abstract

This study delves into the asynchronous filtering problem within discrete-time Takagi–Sugeno (T–S) fuzzy sampled-data systems featuring non-periodic sampled measurements. A non-homogeneous Markov chain is utilized to capture the stochastic nature of the sampling period. The transition probability matrix of this non-homogeneous Markov process is time-varying and characterized using a multi-mode structure. To mitigate conservatism, a quantizer associated with the modes is implemented, where the quantization density is defined as the observed modes. Unlike existing approaches, there exists a discrepancy between the quantizer mode and the sampling mode, following a hidden Markov chain process. Additionally, a novel fuzzy asynchronous filter is introduced based on the observed modes to ensure the stochastic stability of the fuzzy system. Ultimately, the effectiveness of the proposed filter strategy is validated via a tunnel diode circuit model.