Frequency Estimation Using Tapered Data

Author

Quinn, B.G.

Author_Institution

Dept. of Stat., Macquarie Univ., Sydney, NSW

Volume

3

fYear

2006

fDate

14-19 May 2006

Abstract

The maximizer of the periodogram of a sinusoid in additive noise is known to have optimal asymptotic properties even when the noise is neither Gaussian nor white. The effect of tapering or windowing on the accuracy of the estimator does not appear to have been considered previously. In this paper, we present the asymptotic theory for the maximizer of windowed periodograms of Hamming and Hanning-type. We also introduce and analyse two closed-form frequency estimators constructed from three Fourier coefficients of a Hanning-tapered process

Keywords

Fourier series; frequency estimation; spectral analysis; Fourier coefficients; Hamming windowed periodogram; Hanning-tapered process; Hanning-type windowed periodogram; additive noise; asymptotic theory; closed-form frequency estimators; frequency estimation; optimal asymptotic properties; sinusoid periodogram maximizer; tapered data; Additive noise; Estimation theory; Frequency estimation; Gaussian noise; Maximum likelihood estimation; Signal analysis; Signal processing; Statistics; Stochastic processes; Time series analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on

Conference_Location

Toulouse

ISSN

1520-6149

Print_ISBN

1-4244-0469-X

Type

conf

DOI

10.1109/ICASSP.2006.1660593

Filename

1660593