作者: H. Sakai , K. Orita , N. Iwama
DOI: 10.1109/ICASSP.1982.1171708
关键词:
摘要: This paper presents two methods for AR spectrum analysis based on a noisy autocovariance sequence. Unlike most of the traditional methods, we assume that there is an observed sequence, possibly with errors. Such situation occurs in Fourier spectroscopy. We fit models to this sequence by methods. The one least squares autocovariances Yule-Walker equations and other maximum likelihood estimation partial autocorrelation coefficients nonlinear optimization techniques. apply these simulated real plasma data compare their results.