E. Cinquemani, R. Porreca, G. Ferrari-Trecate, and J. Lygeros. A general framework for the identification of jump markov linear systems. Technical Report AUT07-05, Automatic Control Laboratory, ETH Zurich, Switzerland, 2007. http://control.ethz.ch/
This paper concerns parameter identification for the class of jump Markov linear systems. We present recursive formulae for the computation of the second-order statistics of a jump Markov process and discuss their application to the estimation of the model parameters given data from a single or repeated experiments. Simulation results are provided to show the effectiveness of our approach.