DocumentCode
1049756
Title
Bilinear time-frequency representations of signals: the shift-scale invariant class
Author
Hlawatsch, Franz ; Urbanke, Riüdiger L.
Author_Institution
Inst. fur Nachrichtentech. und Hochfrequenztech., Tech. Univ. Wien, Austria
Volume
42
Issue
2
fYear
1994
fDate
2/1/1994 12:00:00 AM
Firstpage
357
Lastpage
366
Abstract
The authors consider the class of bilinear time-frequency representations (BTFR´s) that are invariant (or covariant) to time shifts, frequency shifts, and time-frequency scalings. This “shift-scale invariant” class is the intersection of the classical shift-invariant (Cohen) class and the recently defined affine class. The mathematical description of shift-scale invariant BTFR´s is in terms of a 1-D kernel and is thus particularly simple. The paper concentrates on the time-frequency localization properties of shift-scale invariant BTFR´s. Since any shift-scale invariant BTFR is a superposition of generalized Wigner distributions, the time-frequency localization of the family of generalized Wigner distributions is studied first. For those shift-scale invariant BTFR´s that may be interpreted as smoothed versions of the Wigner distribution (e.g., the Choi-Williams distribution), an analysis in the Fourier transform domain shows interesting peculiarities regarding time-frequency concentration and interference geometry properties
Keywords
Fourier transforms; interference (signal); signal processing; time-frequency analysis; 1-D kernel; Choi-Williams distribution; Fourier transform domain; affine class; bilinear time-frequency representations; frequency shifts; generalized Wigner distributions; interference geometry properties; shift-invariant class; shift-scale invariance; shift-scale invariant smoothing; signal representation; superposition; time shifts; time-frequency concentration; time-frequency localization; time-frequency scalings; Fourier transforms; Frequency domain analysis; Frequency shift keying; Geometry; Interference; Kernel; Physical layer; Spectrogram; Time frequency analysis;
fLanguage
English
Journal_Title
Signal Processing, IEEE Transactions on
Publisher
ieee
ISSN
1053-587X
Type
jour
DOI
10.1109/78.275608
Filename
275608
Link To Document