Title :
Successive correlation based Prony modeling very low SNR behavior analysis
Author :
Gasmi, F. ; Castanie, F.
Author_Institution :
ENSEEIHT-GAPSE, Toulouse, France
Abstract :
The efficiency of the higher-order correlation function as a means to extract sinusoids from white noise with the help of Prony modeling is studied. An improved white-noise-estimation approach is given. It is shown that it is unsatisfactory for low SNR and high order. A statistical analysis is given of the probability of success of the Prony-based modeling of the higher-order correlation. Curves are given to summarize this probability as a function of the SNR
Keywords :
correlation methods; parameter estimation; probability; signal processing; statistical analysis; white noise; Prony modeling; SNR; higher-order correlation function; probability; signal-to-noise ratio; sinusoid extraction; statistical analysis; successive correlation; white noise; white-noise-estimation; Computational modeling; Convergence; Estimation error; Probability; Signal analysis; Signal to noise ratio; Spectral analysis; Statistical analysis; White noise; Writing;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1988. ICASSP-88., 1988 International Conference on
Conference_Location :
New York, NY
DOI :
10.1109/ICASSP.1988.197130
