Fast array algorithm for filtering of Markovian jump linear systems

SUMMARY This paper deals with numerical performance of linear minimum mean square error estimator for discrete-time Markovian jump linear systems. We propose a fast array algorithm to compute this estimator in virtue of the reduction of the computational effort that it provides, if compared with the standard array algorithms. It is particularly useful for this case because the dimension of the linear minimum mean square error estimator increases proportionally to the number of Markovian states. Numerical examples are shown to illustrate the effectiveness of the proposed algorithm. Copyright © 2011 John Wiley & Sons, Ltd.