This article researches the filtering-based parameter estimation issues for a class of multivariate control systems with colored noise. A filtering-based recursive generalized extended least squares algorithm is derived, in which the data filtering technique is used for transforming the original system into two subidentification systems and the least squares principle is used for estimating parameters of these two subsystems. Furthermore, in order to improve the parameter estimation accuracy, the multiinnovation theory is added for deducing a filtering-based multiinnovation recursive generalized extended least squares algorithm.The numerical example confirms that these two proposed algorithms are effective.

No abstract

This paper researches the parameter identification methods of multivariate pseudo-linear moving average systems. A multivariate generalized stochastic gradient (M-GSG) algorithm is presented as a comparison firstly. In order to improve the parameter estimation accuracy, a multivariate multiinnovation generalized stochastic gradient (M-MI-GSG) algorithm and a filtering based multivariate generalized stochastic gradient (F-M-GSG) algorithm are presented by means of the multi-innovation identification theory and the data filtering technique. The simulation results confirm that the proposed algorithms are more effective than the M-GSG algorithm.

No abstract

No abstract