Compressive sensing promises to enable bandwidth-efficient onboard compression of astronomical data by lifting the encoding complexity from the source to the receiver. The signal is recovered off-line, exploiting graphical processing unit (GPU)’s parallel computation capabilities to speedup the reconstruction process. However, inherent GPU hardware constraints limit the size of the recoverable signal and the speedup practically achievable. In this work, we design parallel algorithms that exploit the properties of circulant matrices for efficient GPU-accelerated sparse signals recovery. Our approach reduces the memory requirements, allowing us to recover very large signals with limited memory. In addition, it achieves a 10-fold signal recovery speedup, thanks to adhoc parallelization of matrix–vector multiplications and matrix inversions. Finally, we practically demonstrate our algorithms in a typical application of circulant matrices: deblurring a sparse astronomical image in the compressed domain

GPU-accelerated algorithms for compressed signals recovery with application to astronomical imagery deblurring / Fiandrotti, Attilio; Fosson, Sophie M; Ravazzi, Chiara; Magli, Enrico. - In: INTERNATIONAL JOURNAL OF REMOTE SENSING. - ISSN 0143-1161. - 39:7(2018), pp. 2043-2065. [10.1080/01431161.2017.1356489]

GPU-accelerated algorithms for compressed signals recovery with application to astronomical imagery deblurring

Fiandrotti, Attilio;Fosson, Sophie M;Magli, Enrico
2018

Abstract

Compressive sensing promises to enable bandwidth-efficient onboard compression of astronomical data by lifting the encoding complexity from the source to the receiver. The signal is recovered off-line, exploiting graphical processing unit (GPU)’s parallel computation capabilities to speedup the reconstruction process. However, inherent GPU hardware constraints limit the size of the recoverable signal and the speedup practically achievable. In this work, we design parallel algorithms that exploit the properties of circulant matrices for efficient GPU-accelerated sparse signals recovery. Our approach reduces the memory requirements, allowing us to recover very large signals with limited memory. In addition, it achieves a 10-fold signal recovery speedup, thanks to adhoc parallelization of matrix–vector multiplications and matrix inversions. Finally, we practically demonstrate our algorithms in a typical application of circulant matrices: deblurring a sparse astronomical image in the compressed domain
File in questo prodotto:
File Dimensione Formato  
GPU.pdf

Open Access dal 29/07/2018

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 1.22 MB
Formato Adobe PDF
1.22 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2693597
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo