Title :
High resolution spectrum estimating algorithm
Author :
Luzin, Igor ; Dubinsky, Maxim
fDate :
28 Sep-1 Oct 1998
Abstract :
A high-resolution spectrum estimating algorithm has been developed. The proposed algorithm is based on approximation of maximum likelihood criterion, which leads to two-channel structure of the estimation filter. The first channel corresponds to a measuring channel like one of Capon´s algorithm, and the second one is using for taking into account and compensation of non-coherent white noise component value. The developed algorithm is a compromise solution of the simplicity of DSP-realisation of Capon´s or ARMA-models algorithms and the effectiveness of methods based on covariance matrix singular decomposition
Keywords :
acoustic noise; acoustic signal processing; autoregressive moving average processes; covariance matrices; digital filters; maximum likelihood estimation; singular value decomposition; spectral analysis; white noise; ARMA-models algorithms; Capon´s algorithm; DSP-realisation; covariance matrix singular decomposition; estimation filter; high resolution spectrum estimating algorithm; maximum likelihood; noncoherent white noise component; two-channel structure; Covariance matrix; Eigenvalues and eigenfunctions; Filters; Frequency estimation; Maximum likelihood estimation; Signal processing algorithms; Signal resolution; Spectral analysis; Tuning; White noise;
Conference_Titel :
OCEANS '98 Conference Proceedings
Conference_Location :
Nice
Print_ISBN :
0-7803-5045-6
DOI :
10.1109/OCEANS.1998.726301
