State estimation of discrete-time Markov jump linear systems based on linear minimum mean-square error estimate

This paper considers state estimation problem for discrete-time Markov jump linear systems. For this, two algorithm are presented. The first algorithm is an optimal algorithm of state estimation in the sense of linear minimum mean-square error estimate, which requires an ever-increasing computation and storage load with the length of the noise observation sequence. The second algorithm is a suboptimal algorithm which is proposed to reduce the computation and storage load of the optimal algorithm. A numerical example is presented to evaluate the performance of the proposed suboptimal algorithm.