This paper presents an overview of the methods of array processing as practiced at Floating Point Systems, Inc. The concepts of parallel, pipelined processing - both in hardware and software - are shown to produce very fast computations when applied to highly repetitive calculations. Examples of times for computing 2D FFT's are given.
Published in:
Nuclear Science, IEEE Transactions on
(Volume:26
,
Issue:
2
)
Date of Publication: April 1979