Title :
Recursive identification for multivariate EIV linear systems
Author :
Mu, Bi-Qiang ; Chen, Han-Fu
Author_Institution :
Key Lab. of Syst. & Control, Inst. of Syst. Sci., Beijing, China
Abstract :
The paper concerns identification of the multi-input and multi-output (MIMO) errors-in-variables (EIV) linear systems, for which the errors are described by ARMA processes with unknown coefficients. Recursive estimates are given for coefficients of the linear system, and they are proved to converge to the true values with probability one. A numerical example is given and the simulation results are shown to be consistent with the theoretical analysis.
Keywords :
MIMO systems; autoregressive moving average processes; identification; linear systems; ARMA processes; MIMO; multiinput and multioutput errors-in-variables linear systems; multivariate EIV linear systems; recursive identification; theoretical analysis; Convergence; Least squares approximation; Linear systems; MIMO; Noise; Stochastic processes; Vectors; α-mixing; Errors-in-variables; recursive estimate; stochastic approximation; strong consistency;
Conference_Titel :
Control Conference (CCC), 2012 31st Chinese
Conference_Location :
Hefei
Print_ISBN :
978-1-4673-2581-3
