Accelerating batched 1D-FFT with a CUDA-capable computer

Author

de Beer, R. ; van Ormondt, D. ; Cesare, F. Di ; Graveron-Demilly, D. ; Karras, D.A. ; Starcuk, Z.

Author_Institution

Dept. of Appl. Phys., Univ. of Technol. Delft, Delft, Netherlands

fYear

2010

Firstpage

446

Lastpage

451

Abstract

This work concerns the application of CUDA-based software (Compute Unified Device Architecture), developed by NVIDIA for programmable Graphics Processing units (GPUs). CUDA code is written in ´C for CUDA´, indicating the standard C programming language with NVIDIA extensions.Our goal was to find out, whether batched (multiple) one-dimensional Fast Fourier Transformation (1DFFT), often encountered in various fields of signal processing, can be speeded up significantly by exploiting the parellel-processing power of a low-cost, standard, CUDA-enabled graphics card in a home-assembled PC.

Keywords

Java; computer graphic equipment; coprocessors; fast Fourier transforms; software architecture; software libraries; 1D-FFT; C programming language; CUDA library functions; CUDA-capable computer; GPU; Java environment; NVIDIA; compute unified device architecture; home-assembled PC; one-dimensional Fast Fourier Transformation; programmable graphics processing units; signal processing; Acceleration; Application software; Assembly; Graphics; Hardware; Java; Linux; Magnetic resonance; Software libraries; Software packages; Batched 1D-FFT; CUDA-enabled GPU; CUFFT library; Java bindings; exhaustive search in MRS; home-assembled PC; jMRUI software package;

fLanguage

English

Publisher

ieee

Conference_Titel

Imaging Systems and Techniques (IST), 2010 IEEE International Conference on

Conference_Location

Thessaloniki

Print_ISBN

978-1-4244-6492-0

Type

conf

DOI

10.1109/IST.2010.5548530

Filename

5548530