Implementation of a real-time unfocused SAR algorithm using various computing platforms

This paper presents the comparison of three implementations of an unfocused SAR algorithm in various platform variants. The first presented solution was implemented using G programming language LabVIEW and GPU toolkits for efficient computations on graphical processor. The others two implementations...

Full description

Saved in:
Bibliographic Details
Published in2015 Signal Processing Symposium (SPSympo) pp. 1 - 5
Main Authors Radecki, Krzysztof, Samczynski, Piotr, Kulpa, Krzysztof, Drozdowicz, Jedrzej
Format Conference Proceeding
LanguageEnglish
Published Warsaw University of Technology 01.06.2015
Subjects
CUDA
GPU
Graphics processing units
Imaging
LabView
processing
Radar imaging
Real-time systems
SAR
Spaceborne radar
Synthetic aperture radar
Online AccessGet full text

Cover

More Information
Summary:This paper presents the comparison of three implementations of an unfocused SAR algorithm in various platform variants. The first presented solution was implemented using G programming language LabVIEW and GPU toolkits for efficient computations on graphical processor. The others two implementations were written in C++ with CUDA tools and two different computing platforms. The main aim was to create the implementation capable to perform real-time computations using simplified unfocused SAR algorithm. The presented solutions were tested on real radar data collected during SAR campaign using small aircraft as a radar carrier. This work is a backend to create an UAV-based small and light SAR system with real-time on-board capabilities.
DOI:10.1109/SPS.2015.7168297