This Article 
 Bibliographic References 
 Add to: 
Constant Geometry Fast Fourier Transforms on Array Processors
March 1993 (vol. 42 no. 3)
pp. 371-375

Matrix algebra is used to design and validate parallel algorithms for large constant-geometry fast Fourier transforms (FFTs) on fixed-size array processors. The N-point radix 2 case for a linear array processor with N/2 cells is identical to the usual procedure corresponding to the matrix factorization of M.C. Pease, (1968). The algorithms are engendered by matrix factorizations, which themselves depend on a basic factorization of the perfect shuffle. The resulting data movement is realized in parallel as relatively small perfect shuffles inside each local memory and along each row and column of the array processor, without requiring that the complete array itself have the shuffle-exchange network.

[1] K. E. Batcher, "Sorting networks and their applications," in1968 Spring Joint Comput, Conf., AFIPS Proc., vol. 32, Thompson Books, Washington, DC, 1968, pp. 307-314.
[2] T. Feng, "A survey of interconnection networks,"IEEE Comput. Mag., pp. 12-27, Dec. 1981.
[3] S. W. Golomb, "Permutations by cutting and shuffling,"SIAM Rev., vol. 3, pp. 293-297, 1961.
[4] D. Kleitman, F. T. Leighton, M. Lepley, and G. L. Miller, "New layouts for the shuffle exchange graph," inProc. ACM Symp. Theory Computing, 1981, pp. 278-292.
[5] F. Leighton and G. Miller, "Optimal layouts for small shuffle-exchange graphs,"J. VLSI, pp. 289-299, 1981.
[6] G. Miel, "Constant geometry fast Fourier transforms on array processors," Hughes Research Laboratories Rep., Aug. 6, 1990.
[7] G. Miel, "Trends in systolic and cellular computation,"J. Comput. Appl. Math., vol. 38, pp. 1-25, 1991.
[8] A. Norton and A. J. Silberger, "Parallelization and performance analysis of the Cooley -- Tukey FFT algorithm for shared-memory architectures,"IEEE Trans. Comput., vol. C-36, no. 5, pp. 581-591, May 1987.
[9] M. C. Pease, "An adaptation of the fast Fourier transform for parallel processing,"J. ACM, vol. 15, pp. 252-264, 1968.
[10] M. C. Pease, "The indirect binary n-cube microprocessor array,"IEEE Trans. Comput., vol. C-26, no. 5, pp. 548-573, May 1977.
[11] F. P. Preparata and J. Vuillemin, "The cube-connected cycle: A versatile network for parallel computation,"Commun. ACM, vol. 24, pp. 300-309, May 1981.
[12] H. J. Siegel, "The theory underlying the partitioning of permutation networks,"IEEE Trans. Comput., vol. C-29, no. 9, pp. 791-801, Sept. 1980.
[13] R. C. Singleton, "On computing the fast Fourier transform,"Commun. ACM, vol. 10, pp. 647-654, 1967.
[14] H. S. Stone, "Parallel processing with the perfect shuffle,"IEEE Trans. Comput., vol. C-20, pp. 153-161, Feb. 1971.
[15] Stone, H. S. 1987.High-Performance Computer Architecture. Reading, Mass., Addison-Wesley.
[16] C. D. Thompson, "Fourier transforms in VLSI,"IEEE Trans. Comput., vol. C-32, no. 11, pp. 1047-1057, Nov. 1983.
[17] J. Vuillemin, "A combinatorial limit to the computing power of VLSI circuits," inProc. 21st Symp. Foundations Comput. Sci., IEEE Comput. Soc., Oct. 1980, pp. 94-300.
[18] J. E. Whelchel, J. P. O'Malley, W. J. Rinard, and J. F. McArthur, "The systolic phase rotation FFT--A new algorithm and parallel processor architecture," inProc. 1990 Int. Conf. Acoust., Speech, Signal Processing, IEEE Signal Processing Society, vol. 2, 1990, pp. 1021-1024.
[19] C. Wu and T. Feng, "Universality of the shuffle-exchange network,"IEEE Trans. Comput., vol. C-30, no. 5, pp. 324-331, May 1981.

Index Terms:
constant geometry FFT; fast Fourier transforms; array processors; parallel algorithms; matrix factorization; perfect shuffle; local memory; fast Fourier transforms; logic arrays; matrix algebra; parallel algorithms.
G. Miel, "Constant Geometry Fast Fourier Transforms on Array Processors," IEEE Transactions on Computers, vol. 42, no. 3, pp. 371-375, March 1993, doi:10.1109/12.210180
Usage of this product signifies your acceptance of the Terms of Use.