DocumentCode
2602821
Title
Generalized periodic non-uniform sampling of non-bandlimited signals
Author
Meng, Chen ; Tuqan, Jamal
Author_Institution
Dept. of Electr. & Comput. Eng., California Univ., Davis, CA
fYear
2008
fDate
Jan. 27 2008-Feb. 1 2008
Firstpage
417
Lastpage
421
Abstract
A new paradigm for the periodic non uniform sampling of a class of non bandlimited signals was recently proposed. The main idea is to generate the periodic non uniform samples by retaining a select group of samples from a larger set, obtained by oversampling the continuous time signal by an integer factor L. In this paper, we revisit the problem and consider the more challenging case where the sampler operates at L/M time the Nyquist rate with L and M being coprime integers. We derive a new multi-input multi-output (MIMO) discrete-time model, and then use this model to obtain a set of necessary conditions for perfect signal reconstruction with FIR filters. We also show that unlike the case of M = 1, the newly derived conditions are not sufficient for FIR perfect recovery.
Keywords
FIR filters; signal reconstruction; signal sampling; FIR filters; Nyquist rate; continuous time signal; coprime integers; generalized periodic nonuniform sampling; multi-input multi-output discrete-time model; nonbandlimited signals; perfect signal reconstruction; Digital filters; Finite impulse response filter; IIR filters; Image reconstruction; MIMO; Multiresolution analysis; Sampling methods; Signal generators; Signal reconstruction; Technological innovation; Finite Rate of Innovations; Multirate DSP; Periodic Non-Uniform Sampling; Sylvester; Wavelets;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory and Applications Workshop, 2008
Conference_Location
San Diego, CA
Print_ISBN
978-1-4244-2670-6
Type
conf
DOI
10.1109/ITA.2008.4601083
Filename
4601083
Link To Document