AR, ARMA, and AR-in-Noise Modeling by Fitting Windowed Correlation Data
Document Type
Article
Date of Original Version
1-1-1989
Abstract
A method for autoregressive (AR) modeling of stationary stochastic signals has previously been proposed by the first author [7] based upon leastsquares fitting of the model autocorrelation function (ACF) to the estimated (biased) ACF over more than the minimum number of ACF lags. Here, the method is extended to autoregressive/moving-average (ARMA) models, including the special case of AR signals in white noise, and both AR and ARMA examples are presented. The method differs from the wellknown method of overdetermined normal equations in that fitting error, not equation error, is minimized, and significantly improved performance is obtained. Iterative algorithms patterned after the Steiglitz-McBride deterministic method are derived to solve the resulting nonlinear equations. © 1989 IEEE
Publication Title, e.g., Journal
IEEE Transactions on Acoustics, Speech, and Signal Processing
Volume
37
Issue
10
Citation/Publisher Attribution
Jackson, Leland B., Jianguo Huang, Kevin P. Richards, and Haiguang Chen. "AR, ARMA, and AR-in-Noise Modeling by Fitting Windowed Correlation Data." IEEE Transactions on Acoustics, Speech, and Signal Processing 37, 10 (1989): 1608-1612. doi: 10.1109/29.35404.