TY - GEN

T1 - Unscented compression sensing

AU - Carmi, Avishy

AU - Mihaylova, Lyudmila

AU - Kanevsky, Dimitri

N1 - @inproceedings{DBLP:conf/icassp/CarmiMK12, author = {Avishy Carmi and Lyudmila Mihaylova and Dimitri Kanevsky}, title = {Unscented compressed sensing}, booktitle = {ICASSP}, year = {2012}, pages = {5249-5252}, ee = {http://dx.doi.org/10.1109/ICASSP.2012.6289104}, crossref = {DBLP:conf/icassp/2012}, bibsource = {DBLP, http://dblp.uni-trier.de} }

PY - 2012/3/1

Y1 - 2012/3/1

N2 - In this paper we present a novel compressed sensing (CS) algorithm for the recovery of compressible, possibly time-varying, signal from a sequence of noisy observations. The newly derived scheme is based on the acclaimed unscented Kalman filter (UKF), and is essentially self reliant in the sense that no peripheral optimization or CS algorithm is required for identifying the underlying signal support. Relying exclusively on the UKF formulation, our method facilitates sequential processing of measurements by employing the familiar Kalman filter predictor corrector form. As distinct from other CS methods, and by virtue of its pseudo-measurement mechanism, the CS-UKF, as we termed it, is non iterative, thereby maintaining a computational overhead which is nearly equal to that of the conventional UKF.

AB - In this paper we present a novel compressed sensing (CS) algorithm for the recovery of compressible, possibly time-varying, signal from a sequence of noisy observations. The newly derived scheme is based on the acclaimed unscented Kalman filter (UKF), and is essentially self reliant in the sense that no peripheral optimization or CS algorithm is required for identifying the underlying signal support. Relying exclusively on the UKF formulation, our method facilitates sequential processing of measurements by employing the familiar Kalman filter predictor corrector form. As distinct from other CS methods, and by virtue of its pseudo-measurement mechanism, the CS-UKF, as we termed it, is non iterative, thereby maintaining a computational overhead which is nearly equal to that of the conventional UKF.

KW - Compressed sensing

KW - Kalman filter

KW - Sigma point filter

KW - Sparse signal recovery

KW - Unscented Kalman Filter

U2 - 10.1109/ICASSP.2012.6289104

DO - 10.1109/ICASSP.2012.6289104

M3 - Conference contribution/Paper

SN - 978-1-4673-0045-2

SP - 5249

EP - 5252

BT - Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on

PB - IEEE

CY - Kyoto, Japan

T2 - IEEE Confernce on Acoustics, Speech and Signal Processing (ICASSP)

Y2 - 25 March 2012 through 30 July 2012

ER -