2003
DOI: 10.1016/s0167-8191(03)00040-1
|View full text |Cite
|
Sign up to set email alerts
|

Efficient 2D FFT implementation on mediaprocessors

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2004
2004
2022
2022

Publication Types

Select...
5
2

Relationship

2
5

Authors

Journals

citations
Cited by 18 publications
(4 citation statements)
references
References 26 publications
0
4
0
Order By: Relevance
“…It is observed that the research on parallel FFT is a fast developed field (e.g. see [28][29][30][31], or a relatively complete review in [32]). It is difficult to rank the available FFTs, since there are so many different versions of FFTs, different parallel schemes for the FFTs, and different parallel computers to implement them.…”
Section: A Brief Discussion For the Pfft Schemementioning
confidence: 99%
“…It is observed that the research on parallel FFT is a fast developed field (e.g. see [28][29][30][31], or a relatively complete review in [32]). It is difficult to rank the available FFTs, since there are so many different versions of FFTs, different parallel schemes for the FFTs, and different parallel computers to implement them.…”
Section: A Brief Discussion For the Pfft Schemementioning
confidence: 99%
“…INSPIRE recursively adopts a 2D NTT algorithm to achieve such mapping [41]. The key idea of 2D NTT is that we can break down a length-𝐿 = 𝑚 × 𝑛 NTT into smaller length NTTs, by performing length-𝑚 NTT 𝑛 times and performing length-𝑛 NTT 𝑚 times.…”
Section: Recursive Ntt Unitmentioning
confidence: 99%
“…By contrast, we could rapidly obtain the output responses in another way, which avoids solving equations directly. In that way, we firstly sample points in the output responses, and then recover the output responses through fast Fourier transform (FFT) [6][7][8]. The FFT plays a significant role in signal processing and recovering.…”
Section: Introductionmentioning
confidence: 99%