Two-dimensional systolic-array architecture for pixel-level vision tasks

2010 ◽  
Author(s):  
Julien A. Vijverberg ◽  
Peter H. N. de With
1994 ◽  
Vol 20 (3) ◽  
pp. 233-241
Author(s):  
Mansour H. Jaragh ◽  
Malek G.M. Hussain

1987 ◽  
Vol 12 (11) ◽  
pp. 959 ◽  
Author(s):  
M. Shabeer ◽  
I. Andonovic ◽  
B. Culshaw

1997 ◽  
Vol 33 (24) ◽  
pp. 2055 ◽  
Author(s):  
H. Amin ◽  
K.M. Curtis ◽  
B.R. Hayes-Gill

1992 ◽  
Vol 139 (2) ◽  
pp. 147
Author(s):  
G.A. Orton ◽  
L.E. Peppard ◽  
S.G. Akl

1997 ◽  
Vol 08 (06) ◽  
pp. 1317-1334 ◽  
Author(s):  
Thomas Lippert ◽  
Klaus Schilling ◽  
Sven Trentmann ◽  
Federico Toschi ◽  
Raffaele Tripiccione

We present a parallel FFT algorithm for SIMD systems following the "Transpose Algorithm" approach. The method is based on the assignment of the data field onto a one-dimensional ring of systolic cells. The systolic array can be universally mapped onto any parallel system. In particular for systems with next-neighbor connectivity our method has the potential to improve the efficiency of matrix transposition by use of hyper-systolic communication. We have realized a scalable parallel FFT on the APE100/Quadrics massively parallel computer, where our implementation is part of a two-dimensional hydrodynamics code for turbulence studies.


Sign in / Sign up

Export Citation Format

Share Document