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.
Andrei Variu, Cheng Zhao, Yu Yu, Hanyu Zhang
Jan Skaloud, Davide Antonio Cucci, Kenneth Joseph Paul