Title :
Multiple-GPU accelerated range-Doppler algorithm for synthetic aperture radar imaging
Author :
Ning, Xia ; Yeh, Chunmao ; Zhou, Bin ; Gao, Wei ; Yang, Jian
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
Abstract :
Parallelization based on general purpose graphic processing unit (GPGPU) is a kind of promising technology widely used to improve the computing performance. GPU is designed for data parallel and computational intensive applications. In this paper, a d ata processor based on multiple GPGPU devices is propo sed for SAR signal processing. Some key technologies are emphasized for exploiting the best performance of GPGPU. And t he processor´s performance is demonstrated by the implementation of range-Doppler imaging algorithm, which shows it is suitable for real-time SAR signal processing.
Keywords :
computer graphic equipment; coprocessors; radar imaging; synthetic aperture radar; GPGPU devices; computational intensive application; data parallel application; multiple-GPU accelerated range-doppler imaging; real-time SAR signal processing; synthetic aperture radar imaging; Algorithm design and analysis; Apertures; Graphics processing unit; Instruction sets; Performance evaluation; Signal processing algorithms; Synthetic aperture radar;
Conference_Titel :
Radar Conference (RADAR), 2011 IEEE
Conference_Location :
Kansas City, MO
Print_ISBN :
978-1-4244-8901-5
DOI :
10.1109/RADAR.2011.5960627
