Energy-to-peak Mode-dependent Filter Design for Discrete-time Markovian Jump Linear Systems with Intermittent Measurements

摘要

In this paper, we investigate the robust energy-to-peak filtering for discrete-time Markovian jump linear systems with intermittent measurements. The intermittent measurements are modeled as a Bernoulli process with a fixed probability. By defining the filtering error, an augmented estimation error system is obtained. Due to the existence of the stochastic variables, the exponentially mean-square stability and the robust energy-to-peak performance are both studied for the estimation error system. The widely-used Lyapunov theory is employed to investigate the stability and the energy-to-peak performance. Sufficient conditions expressed in the form of matrix inequalities are obtained. Since the estimator parameters and the Lyapunov weighting matrix to be determined are coupled with each other, a partition technique is utilized to decouple the estimator parameters and the Lyapunov weighting matrix. After defining several new parameters, we develop the design approach for the mode-dependent energy-to-peak estimator. The estimator parameters can be derived by solving a set of linear matrix inequalities. Finally, a numerical example is used to show the effectiveness of the proposed design approach.