An Architecture for a Video Rate Two-Dimensional Fast Fourier Transform Processor

A description of an architecture capable of computing two-dimensional fast Fourier transforms on a 256*256 pixel image at a rate of 30 images per second is presented. The architecture consists of a small number of basic building blocks which may be repeated to yield any desired performance. To achieve video rate performance, 16 butterfly processors, arranged as four coupled clusters of four processors each, and nine working memories are required. Because of the parallelism and pipelining used in the design, the system clock needed to achieve this high level of performance is only 240 ns. >

[1]  Kai Hwang,et al.  PUMPS Architecture for Pattern Analysis and Image Database Management , 1982, IEEE Transactions on Computers.

[2]  Michael J. Corinthios A Fast Fourier Transform for High-Speed Signal Processing , 1971, IEEE Transactions on Computers.

[3]  Randy H. Steinvorth,et al.  A video rate two dimensional FFT processor , 1980, ICASSP.

[4]  Kenneth C. Smith,et al.  A Parallel Radix-4 Fast Fourier Transform Computer , 1975, IEEE Transactions on Computers.

[5]  Earl E. Swartzlander,et al.  Computer Arithmetic , 1980 .

[6]  Azriel Rosenfeld,et al.  Image Processing on ZMOB , 1982, IEEE Transactions on Computers.

[7]  Earl E. Swartzlander,et al.  FREQUENCY-DOMAIN DIGITAL FILTERING WITH VLSI. , 1985 .

[8]  Michael J. Corinthios,et al.  The Design of a Class of Fast Fourier Transform Computers , 1971, IEEE Transactions on Computers.

[9]  Marshall C. Pease,et al.  An Adaptation of the Fast Fourier Transform for Parallel Processing , 1968, JACM.

[10]  C. K. Yuen,et al.  Theory and Application of Digital Signal Processing , 1978, IEEE Transactions on Systems, Man, and Cybernetics.