Title :
A two-step method for nonminimum phase ARMA identification
Author :
Fu, Zhi-Zhen ; Paarmann, Larry D. ; Korenberg, Michael J.
Author_Institution :
Dept. of Electr. Eng., Wichita State Univ., KS, USA
Abstract :
An identification method to estimate the unknown parameters of a time-invariant, possibly nonminimum-phase ARMA (autoregressive moving average) system with inaccessible input is considered as a two-step procedure. In the first step, a spectrally equivalent minimum-phase system is estimated by using an orthogonal least-squares method. In the second step, D spectrally equivalent ARMA systems are considered, where D is the number of phase realizations. A decision criterion is then applied to determine which phase realization is the correct one
Keywords :
filtering and prediction theory; parameter estimation; signal processing; decision criterion; nonminimum phase ARMA identification; orthogonal least-squares method; phase realizations; spectrally equivalent minimum-phase system; two-step method; unknown parameters; Delay estimation; Gaussian processes; Parameter estimation; Phase estimation; Predictive models; Signal processing; State estimation; Statistics; Stochastic systems; System identification;
Conference_Titel :
Circuits and Systems, 1991., Proceedings of the 34th Midwest Symposium on
Conference_Location :
Monterey, CA
Print_ISBN :
0-7803-0620-1
DOI :
10.1109/MWSCAS.1991.252136
