Title :
Computation of autoregressive model coefficients through orthogonal projection
Author :
Ning, Taikang ; Huang, Wenli
Author_Institution :
Dept. of Eng., Trinity Coll., Hartford, CT, USA
Abstract :
An autoregressive (AR) modeling technique based on Gram-Schmidt orthogonal transform (GSOT) is introduced. The GSOT method is computation efficient and provides accurate estimation of the AR model coefficients and power spectra. The algorithm of the GSOT method is outlined and, with extensive simulation tests, its superior performance is justified which compared to the popular Burg (1968) method
Keywords :
autoregressive processes; parameter estimation; spectral analysis; transforms; AR modeling; Burg method; Gram-Schmidt orthogonal transform; autoregressive model coefficients; computation efficiency; orthogonal projection; performance; power spectrum estimation; simulation tests; Computational modeling; Constraint optimization; Ear; Frequency estimation; Least squares methods; Power engineering and energy; Power generation; Reflection; Signal to noise ratio; Testing;
Conference_Titel :
Signal Processing, 1996., 3rd International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-2912-0
DOI :
10.1109/ICSIGP.1996.567096
