2015 16th International Conference on Computational Problems of Electrical Engineering (CPEE) 2015
DOI: 10.1109/cpee.2015.7333365
|View full text |Cite
|
Sign up to set email alerts
|

Effectiveness of Fast Fourier Transform implementations on GPU and CPU

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2018
2018
2021
2021

Publication Types

Select...
4
1
1

Relationship

1
5

Authors

Journals

citations
Cited by 7 publications
(5 citation statements)
references
References 5 publications
0
5
0
Order By: Relevance
“…The data-independent parameters are: the definition of interpolation node indices, interpolation weights, and a filter matrix for combined processing of focal plane selection; compensation of limited angle detection; and Hilbert transform. The implemented pipeline consists of the simplest arithmetic operations with the exception of forward/inverse FFT 2D functions (Figure 2) [14]. The number of spatial frequencies ( , ) is chosen to be identical to the number of pixels × .…”
Section: Gpu-fd Accelerated Reconstructionmentioning
confidence: 99%
See 1 more Smart Citation
“…The data-independent parameters are: the definition of interpolation node indices, interpolation weights, and a filter matrix for combined processing of focal plane selection; compensation of limited angle detection; and Hilbert transform. The implemented pipeline consists of the simplest arithmetic operations with the exception of forward/inverse FFT 2D functions (Figure 2) [14]. The number of spatial frequencies ( , ) is chosen to be identical to the number of pixels × .…”
Section: Gpu-fd Accelerated Reconstructionmentioning
confidence: 99%
“…For enhanced reconstruction quality within the depth range , it is also important that the acquisition range is wider than , so that the diagonals corresponding to the maximum receiving angles of the detector are fully covered. However, for simplicity we will assume that the The implemented pipeline consists of the simplest arithmetic operations with the exception of forward/inverse FFT 2D functions (Figure 2) [14]. The number of spatial frequencies (k X , k Z ) is chosen to be identical to the number of pixels N X × N Z .…”
Section: Gpu-fd Accelerated Reconstructionmentioning
confidence: 99%
“…Discrete Fourier transforms of uniformly spaced inputs are often computed with a fast Fourier transform (FFT) algorithm, which can be parallelized for CPUs 1099 or GPUs 65,[1100][1101][1102] . Typically, the speedup of FFTs on GPUs over CPUs is higher for larger signals 1103,1104 . Most popular FFTs are based on the Cooley-Turkey algorithm 1105,1106 , which recursively divides FFTs into smaller FFTs.…”
Section: /98mentioning
confidence: 99%
“…It should be noted that fast parametric transforms with such computational structures enable their effective implementations in parallel or massively-parallel approaches using graphics processing units (GPU) (c.f. [17,18]). The GPU accelerated algorithms constitute dynamically developing branch of computational techniques used extensively in many practical engineering problems (see e.g.…”
Section: Fast Parametric Transformsmentioning
confidence: 99%