DocumentCode :
3004092
Title :
A lattice algorithm for Lp deconvolution
Author :
Alcázar-Fernández, J. ; Fraile-Peláez, F.J.
Author_Institution :
ETSI Telecomunication, Univ. Politecnica de Madrid, Spain
fYear :
1988
fDate :
11-14 Apr 1988
Firstpage :
1746
Abstract :
Predictive deconvolution is traditionally based on a least-squares error criterion (L2 norm minimization). The authors propose a fast and stable algorithm for Lp, 1⩽ p⩽2, that is based on a lattice structure. The lattice coefficients are obtained using a Burg algorithm with the Lp criterion; that is, at every cell the reflection coefficient which minimizes the sum of the Lp norms of the forward and backward prediction errors at the output of the cell is chosen
Keywords :
filtering and prediction theory; least squares approximations; Burg algorithm; Lp deconvolution; lattice algorithm; least-squares error criterion; predictive deconvolution; reflection coefficient; Deconvolution; Iterative algorithms; Lattices; Least squares methods; Nonlinear filters; Reflection; Robustness; Stability; Telecommunications; Transversal filters;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1988. ICASSP-88., 1988 International Conference on
Conference_Location :
New York, NY
ISSN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.1988.196956
Filename :
196956
Link To Document :
https://search.ricest.ac.ir/dl/search/defaultta.aspx?DTC=49&DC=3004092
بازگشت