Wavelet design of time-varying filters

Author

Ridsdill-Smith, Thomas A.

Author_Institution

Dept. of Geol. & Geophys., Western Australia Univ., Nedlands, WA, Australia

Volume

2

fYear

1999

fDate

1999

Firstpage

599

Abstract

Linear time-varying filters can be efficiently implemented using a construction based on the wavelet transform. The method generates a vector of weights for the wavelet coefficients, known as the scale filter coefficients, using least-squares techniques. By sampling the wavelet filters and the desired transfer function of the time-varying filters using an exponential sampling comb, the least-squares inversion can be made stable and fast. Scale filters are used in aeromagnetic data processing to correct for survey altitude effects where they demonstrate a major improvement over conventional time-domain filters, especially in the low-frequency response

Keywords

filtering theory; least squares approximations; signal sampling; time-frequency analysis; time-varying filters; transfer functions; wavelet transforms; aeromagnetic data processing; exponential sampling comb; filter design; least-squares techniques; linear time-varying filters; low-frequency response; scale filter coefficients; survey altitude effects; transfer function; wavelet filters; wavelet transform; Continuous wavelet transforms; Convolution; Filtering; Fourier transforms; Nonlinear filters; Sampling methods; Signal processing; Transfer functions; Wavelet domain; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on

Conference_Location

Brisbane, Qld.

Print_ISBN

1-86435-451-8

Type

conf

DOI

10.1109/ISSPA.1999.815744

Filename

815744