This paper describes techniques for reducing CPU time, memory, and I/O time required to perform a three‐dimensional convolution on real data, and the use of an array processor. These techniques capitalize on the Hermitian nature of the Fourier transform of mathematically real input data, take advantage of convolution kernels having mirror symmetry in three dimensions, and eliminate trivial transformations. The result is convolution over a three‐dimensional 64×64×64 array containing 250 K real elements that requires only 28 s to execute. Memory space of a medium‐sized array processor is sufficiently large for the procedures to be carried out on entire real, single‐precision arrays at one time without transferring data between the array processor and disk storage.
Skip Nav Destination
Article navigation
Research Article|
September 01 1990
Three‐dimensional Fourier convolution with an array processor Free
N. H. Wells;
N. H. Wells
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Search for other works by this author on:
C. S. Burrus;
C. S. Burrus
Department of Electrical and Computer Engineering, Rice University, Houston, Texas 77251
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Search for other works by this author on:
G. E. Desobry;
G. E. Desobry
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Search for other works by this author on:
A. L. Boyer
A. L. Boyer
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Search for other works by this author on:
N. H. Wells
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
C. S. Burrus
Department of Electrical and Computer Engineering, Rice University, Houston, Texas 77251
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
G. E. Desobry
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
A. L. Boyer
Department of Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Comput. Phys. 4, 507–513 (1990)
Article history
Received:
November 06 1989
Accepted:
May 18 1990
Citation
N. H. Wells, C. S. Burrus, G. E. Desobry, A. L. Boyer; Three‐dimensional Fourier convolution with an array processor. Comput. Phys. 1 September 1990; 4 (5): 507–513. https://doi.org/10.1063/1.168385
Download citation file:
Citing articles via
Related Content
The TRIMM parallel processor
Comput. Phys. (July 1991)
Operating experience with a multiple‐processor system for data acquisition and reduction in nuclear physics
Comput. Phys. (September 1989)
Analysis of high‐perveance electron guns
Comput. Phys. (May 1990)
The Polyp multiprocessor: Architecture and applications in nuclear physics
Comput. Phys. (May 1990)
Implementation of the Lanczos algorithm for the Hubbard model on the Connection Machine system
Comput. Phys. (November 1992)