Processing of synthetic aperture radar data with GPGPU

Carmine Clemente, Maurizio Di Bisceglie, Michele Di Santo, Nadia Ranaldo, Marcello Spinelli

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

22 Citations (Scopus)

Abstract

Synthetic aperture radar processing is a complex task that involves advanced signal processing techniques and intense computational effort. While the first issue has now reached a mature stage, the question of how to produce accurately focused images in real-time, without mainframe facilities, is still under debate. The recent introduction of general-purpose graphic processing units seems to be quite promising in this view, especially for the decreased per-core cost barrier and for the affordable programming complexity. The authors explain, in this work, the main computational features of a range-Doppler Synthetic Aperture Radar (SAR) processor, trying to disclose the degree of parallelism in the operations at the light of the CUDA programming model. Given the extremely flexible structure of the Single Instruction Multiple Threads (SIMT) model, the authors show that the optimization of a SAR processing unit cannot reduce to an FFT optimization, although this is a quite extensively used kernel. Actually, it is noticeable that the most significant advantage is obtained in the range cell migration correction kernel where a complex interpolation stage is performed very efficiently exploiting the SIMT model. Performance show that, using a single Nvidia Tesla-C1060 GPU board, the obtained processing time is more than fifteen time better than our test workstation.
LanguageEnglish
Title of host publicationIEEE Workshop on Signal Processing Systems, 2009
Subtitle of host publicationSiPS 2009
PublisherIEEE
Pages309-314
Number of pages6
ISBN (Print)978-1-4244-4335-2
DOIs
Publication statusPublished - Oct 2009
EventIEEE Workshop on Signal Processing Systems, 2009. SiPS 2009. - Tampere, Finland
Duration: 7 Oct 200910 Oct 2009

Conference

ConferenceIEEE Workshop on Signal Processing Systems, 2009. SiPS 2009.
CountryFinland
CityTampere
Period7/10/0910/10/09

Fingerprint

Synthetic aperture radar
Processing
Doppler radar
Flexible structures
Fast Fourier transforms
Interpolation
Signal processing
Costs
Graphics processing unit

Keywords

  • processing
  • synthetic aperture
  • radar data
  • gpgpu

Cite this

Clemente, C., Di Bisceglie, M., Di Santo, M., Ranaldo, N., & Spinelli, M. (2009). Processing of synthetic aperture radar data with GPGPU. In IEEE Workshop on Signal Processing Systems, 2009: SiPS 2009 (pp. 309-314). IEEE. https://doi.org/10.1109/SIPS.2009.5336272
Clemente, Carmine ; Di Bisceglie, Maurizio ; Di Santo, Michele ; Ranaldo, Nadia ; Spinelli, Marcello. / Processing of synthetic aperture radar data with GPGPU. IEEE Workshop on Signal Processing Systems, 2009: SiPS 2009. IEEE, 2009. pp. 309-314
@inproceedings{b0fbad62bb3044988103e81b11ee64cf,
title = "Processing of synthetic aperture radar data with GPGPU",
abstract = "Synthetic aperture radar processing is a complex task that involves advanced signal processing techniques and intense computational effort. While the first issue has now reached a mature stage, the question of how to produce accurately focused images in real-time, without mainframe facilities, is still under debate. The recent introduction of general-purpose graphic processing units seems to be quite promising in this view, especially for the decreased per-core cost barrier and for the affordable programming complexity. The authors explain, in this work, the main computational features of a range-Doppler Synthetic Aperture Radar (SAR) processor, trying to disclose the degree of parallelism in the operations at the light of the CUDA programming model. Given the extremely flexible structure of the Single Instruction Multiple Threads (SIMT) model, the authors show that the optimization of a SAR processing unit cannot reduce to an FFT optimization, although this is a quite extensively used kernel. Actually, it is noticeable that the most significant advantage is obtained in the range cell migration correction kernel where a complex interpolation stage is performed very efficiently exploiting the SIMT model. Performance show that, using a single Nvidia Tesla-C1060 GPU board, the obtained processing time is more than fifteen time better than our test workstation.",
keywords = "processing, synthetic aperture, radar data, gpgpu",
author = "Carmine Clemente and {Di Bisceglie}, Maurizio and {Di Santo}, Michele and Nadia Ranaldo and Marcello Spinelli",
year = "2009",
month = "10",
doi = "10.1109/SIPS.2009.5336272",
language = "English",
isbn = "978-1-4244-4335-2",
pages = "309--314",
booktitle = "IEEE Workshop on Signal Processing Systems, 2009",
publisher = "IEEE",

}

Clemente, C, Di Bisceglie, M, Di Santo, M, Ranaldo, N & Spinelli, M 2009, Processing of synthetic aperture radar data with GPGPU. in IEEE Workshop on Signal Processing Systems, 2009: SiPS 2009. IEEE, pp. 309-314, IEEE Workshop on Signal Processing Systems, 2009. SiPS 2009. , Tampere, Finland, 7/10/09. https://doi.org/10.1109/SIPS.2009.5336272

Processing of synthetic aperture radar data with GPGPU. / Clemente, Carmine; Di Bisceglie, Maurizio; Di Santo, Michele; Ranaldo, Nadia; Spinelli, Marcello.

IEEE Workshop on Signal Processing Systems, 2009: SiPS 2009. IEEE, 2009. p. 309-314.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Processing of synthetic aperture radar data with GPGPU

AU - Clemente, Carmine

AU - Di Bisceglie, Maurizio

AU - Di Santo, Michele

AU - Ranaldo, Nadia

AU - Spinelli, Marcello

PY - 2009/10

Y1 - 2009/10

N2 - Synthetic aperture radar processing is a complex task that involves advanced signal processing techniques and intense computational effort. While the first issue has now reached a mature stage, the question of how to produce accurately focused images in real-time, without mainframe facilities, is still under debate. The recent introduction of general-purpose graphic processing units seems to be quite promising in this view, especially for the decreased per-core cost barrier and for the affordable programming complexity. The authors explain, in this work, the main computational features of a range-Doppler Synthetic Aperture Radar (SAR) processor, trying to disclose the degree of parallelism in the operations at the light of the CUDA programming model. Given the extremely flexible structure of the Single Instruction Multiple Threads (SIMT) model, the authors show that the optimization of a SAR processing unit cannot reduce to an FFT optimization, although this is a quite extensively used kernel. Actually, it is noticeable that the most significant advantage is obtained in the range cell migration correction kernel where a complex interpolation stage is performed very efficiently exploiting the SIMT model. Performance show that, using a single Nvidia Tesla-C1060 GPU board, the obtained processing time is more than fifteen time better than our test workstation.

AB - Synthetic aperture radar processing is a complex task that involves advanced signal processing techniques and intense computational effort. While the first issue has now reached a mature stage, the question of how to produce accurately focused images in real-time, without mainframe facilities, is still under debate. The recent introduction of general-purpose graphic processing units seems to be quite promising in this view, especially for the decreased per-core cost barrier and for the affordable programming complexity. The authors explain, in this work, the main computational features of a range-Doppler Synthetic Aperture Radar (SAR) processor, trying to disclose the degree of parallelism in the operations at the light of the CUDA programming model. Given the extremely flexible structure of the Single Instruction Multiple Threads (SIMT) model, the authors show that the optimization of a SAR processing unit cannot reduce to an FFT optimization, although this is a quite extensively used kernel. Actually, it is noticeable that the most significant advantage is obtained in the range cell migration correction kernel where a complex interpolation stage is performed very efficiently exploiting the SIMT model. Performance show that, using a single Nvidia Tesla-C1060 GPU board, the obtained processing time is more than fifteen time better than our test workstation.

KW - processing

KW - synthetic aperture

KW - radar data

KW - gpgpu

UR - http://www.scopus.com/inward/record.url?scp=74549177009&partnerID=8YFLogxK

U2 - 10.1109/SIPS.2009.5336272

DO - 10.1109/SIPS.2009.5336272

M3 - Conference contribution book

SN - 978-1-4244-4335-2

SP - 309

EP - 314

BT - IEEE Workshop on Signal Processing Systems, 2009

PB - IEEE

ER -

Clemente C, Di Bisceglie M, Di Santo M, Ranaldo N, Spinelli M. Processing of synthetic aperture radar data with GPGPU. In IEEE Workshop on Signal Processing Systems, 2009: SiPS 2009. IEEE. 2009. p. 309-314 https://doi.org/10.1109/SIPS.2009.5336272