DocumentCode :
1016135
Title :
Shear madness: new orthonormal bases and frames using chirp functions
Author :
Baraniuk, Richard G. ; Jones, Douglas L.
Author_Institution :
Dept. of Electr. & Comput. Eng., Rice Univ., Houston, TX, USA
Volume :
41
Issue :
12
fYear :
1993
fDate :
12/1/1993 12:00:00 AM
Firstpage :
3543
Lastpage :
3549
Abstract :
The proportional-bandwidth and constant-bandwidth time-frequency signal decompositions of the wavelet, Gabor, and Wilson orthonormal bases have attracted substantial interest for representing nonstationary signals. However, these representations are limited in that they are based on rectangular tessellations of the time-frequency plane. While much effort has gone into methods for designing nice wavelet and window functions for these frameworks, little consideration has been given to methods for constructing orthonormal bases employing nonrectangular time-frequency tilings. The authors take a first step in this direction by deriving two new families of orthonormal bases and frames employing elements that shear, or chirp, in the time-frequency plane, in addition to translate and scale. The new scale-shear fan bases and shift-shear chevron bases are obtained by operating on an existing: wavelet, Gabor (1946), or Wilson basis set with two special unitary warping transformations. In addition to the theoretical benefit of broadening the class of valid time-frequency plane tilings, these new bases could possibly also be useful for representing certain types of signals, such as chirping and dispersed signals
Keywords :
signal processing; time-frequency analysis; wavelet transforms; Gabor bases; Wilson bases; chirp functions; constant bandwidth decomposition; dispersed signals; frames; nonrectangular time-frequency tilings; nonstationary signals; orthonormal bases; proportional bandwidth decomposition; scale-shear fan bases; shift-shear chevron bases; time-frequency plane; time-frequency signal decompositions; unitary warping transformations; wavelet; wavelet functions; window functions; Chirp; Computer science education; Continuous wavelet transforms; Design methodology; Educational programs; Fourier transforms; Multidimensional systems; Signal resolution; Time frequency analysis; Wavelet transforms;
fLanguage :
English
Journal_Title :
Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1053-587X
Type :
jour
DOI :
10.1109/78.258094
Filename :
258094
Link To Document :
بازگشت