Title :
Estimation of Fourier coefficients
Author :
den Bos, Adriaan Van
Author_Institution :
Dept. of Appl. Phys., Delft Univ. of Technol., Netherlands
fDate :
10/1/1989 12:00:00 AM
Abstract :
It is shown that the maximum-likelihood estimation or robust estimation of the Fourier coefficients may be preferable to Fourier transformation if the noise contains outliers or is otherwise not normally distributed. The reason is that, in that case, these estimators produce Fourier coefficient estimates and, therefore, system parameter estimates having a smaller variance
Keywords :
Fourier transforms; parameter estimation; spectral analysis; Fourier coefficients; maximum-likelihood estimation; robust estimation; spectral analysis; Fourier transforms; Genetic expression; Least squares approximation; Least squares methods; Maximum likelihood estimation; Noise reduction; Noise robustness; Parameter estimation; Physics; Stochastic processes;
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
