Signal Reconstruction via Compressive Sensing (CROSBI ID 575558)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Tralić, Dijana ; Grgić, Sonja
engleski
Signal Reconstruction via Compressive Sensing
Compressive sensing is a new approach of sampling theory, which assumes that signal can be exactly recovered from incomplete information. It relies on properties such as incoherence, signal sparsity and compressibility, and does not follow traditional acquisition process based on transform coding. Sensing procedure is very simple, nonadaptive method that employs linear projections of signal onto test functions. Set of test functions is arranged in the measurement matrix that allows acquiring random samples of original signal. Signal reconstruction is achieved from small amount of data by an optimization process which has the aim to find the sparsest vector with transform coefficients among all possible solutions. This paper gives an overview of compressive sensing theory, background, measurement and reconstruction processes. Reconstruction process was presented on a few types of signals at the end of this paper. Experimental results show that accurate reconstruction is possible for various type of signals.
Compressive Sensing; Compressive Sampling; Sparsity; Compressibility; Random Projections; Nonlinear Reconstruction
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
5-9.
2011.
objavljeno
Podaci o matičnoj publikaciji
Proceedings of the 53rd International Symposium ELMAR-2011
Božek, Jelena ; Grgić, Mislav
Zagreb: Hrvatsko društvo Elektronika u pomorstvu (ELMAR)
978-953-7044-12-1
1334-2630
Podaci o skupu
International Symposium ELMAR
predavanje
14.09.2011-16.09.2011
Zadar, Hrvatska