Title :
A novel recursive approach to estimating MA parameters of causal ARMA models from cumulants
Author :
Zhang, Xian-Da ; Zhou, Yanli
Author_Institution :
Dept. of Electr. & Comput. Eng., California Univ., La Jolla, CA, USA
fDate :
11/1/1992 12:00:00 AM
Abstract :
The estimation of MA (moving average) parameters of causal ARMA (autoregressive moving average) models using the third-order cumulants alone is treated. A novel recursive approach is proposed. Under certain conditions, the approach applies to noncausal models
Keywords :
parameter estimation; signal processing; additive Gaussian coloured noise; autoregressive moving average; causal ARMA models; moving average parameters estimation; noncausal models; recursive approach; third-order cumulants; Adaptive filters; Additive noise; Colored noise; Filtering theory; Gaussian noise; IIR filters; Parameter estimation; Recursive estimation; Signal processing algorithms; Speech;
Journal_Title :
Signal Processing, IEEE Transactions on
