• Publication
  • 1994
  • Issue No. 4 - April
  • Abstract - Optimal Parallel and Pipelined Processing Through a New Class of Matrices with Application to Generalized Spectral Analysis
 This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Optimal Parallel and Pipelined Processing Through a New Class of Matrices with Application to Generalized Spectral Analysis
April 1994 (vol. 43 no. 4)
pp. 443-459

A new class of general-base matrices, named sampling matrices, which are meant to bridge the gap between algorithmic description and computer architecture is proposed. "Poles," "zeros," "pointers," and "spans" are among the terms introduced to characterize properties of this class of matrices. A formalism for the decomposition of a general matrix in terms of general-base sampling matrices is proposed. "Span" matrices are introduced to measure the dependence of a matrix span on algorithm parameters and, among others, the interaction between this class of matrices and the general-base perfect shuffle permutation matrix previously introduced. A classification of general-base parallel "recirculant" and parallel pipelined processors based on memory topology, access uniformity and shuffle complexity is proposed. The matrix formalism is then used to guide the search for algorithm factorizations leading to optimal parallel and pipelined processor architecture.

[1] Y.-Y. Jou and J. A. Abraham, "Fault tolerant FFT networks,"IEEE Trans. Comput., vol. 37, no. 5, May 1988, pp. 548-561.
[2] H. K. Nagpal, G. A. Jullien, and W. C. Miller, "Processor architectures for two-dimensional convolvers using a single multiplexed computational element with finite field arithmetic,"IEEE Trans. Comput., vol. C-32, no. 11, pp. 989-1001, Nov, 1983.
[3] G. F. Taylor, R. H. Steinvorth, and J. F. McDonald, "An architecture for a video rate two-dimensional fast fourier transform processor,"IEEE Trans. Comput, vol. 37, no. 9, pp. 1145-1151, Sept. 1988.
[4] C. Moraga, "Design of a multiple-valued systolic system for the computation of the chrestenson spectrum,"IEEE Trans. Comput., vol. C-35, no. 2, pp. 183-188, Feb. 1986.
[5] C. F. Chen and W. K. Leung, "Algorithms for converting sequency, dyadic and Hadamard ordered Walsh functions,"Math and Comput. in Sim., vol. 27, pp. 471-478, 1985.
[6] P. M. Grant and J. H. Collins, "Introduction to electronic warfare,"IEEE Proc., vol. 129, pt. F, no. 3, pp. 113-132, June 1982.
[7] T. D. Roziner, M. G. Karpovsky, and L. A. Trachtenberg, "Fast Fourier transforms over finite groups by multiprocessor systems,"IEEE Trans. Accoustics, Speech&Signal Processing, vol. 38, no. 2, pp. 226-240, Feb. 1990.
[8] C. Moraga and K. Seseke, "The Chrestenson transform in pattern analysis," inSpectral Techniques and Fault Detection, M. Karpovsky, Ed. New York: Academic Press, 1985.
[9] M. J. Corinthios, "3-D cellular arrays for parallel/cascade image/signal processing," inSpectral Techniques and Fault Detection, M. Karpovsky, Ed. New York: Academic Press, 1985.
[10] M. J. Corinthios, "Pipe-line constant geometry algorithms and processors for the generalized spectral analysis of images," presented at theProc. IMACSIFAC Symp. Parallel and Distrib. Computing--P.D. COM'91, Corfu, Greece, June 23-28, 1991.
[11] H. E. Chrestenson, "A class of generalized Walsh functions,"Pacific J. of Math., vol. 5, pp. 17-31, 1955.
[12] M. J. Corinthios, "The design of a class of fast Fourier transform computers,"IEEE Trans. Comput., vol. C-20, pp. 617-623, June 1971.
[13] H. Sloate, "Matrix representations for sorting and the fast Fourier transform,"IEEE Trans. Circuits&Syst., vol. CAS-21, no. 1, pp. 109-116, Jan. 1974.
[14] N. A. Alexandridis, "Relations among sequency, axis symmetry and period of Walsh functions,"IEEE Trans. Inform. Theory, pp. 495-497, July 1971.
[15] Y. A. Geadah and M. J. Corinthios, "Natural, dyadic and sequency algorithms and processors for the Walsh-Hadamard transform,"IEEE Trans. Comput., vol. C-26, no. 5, pp. 435-442, May 1977.
[16] M. J. Corinthios, K. C. Smith, and J. L. Yen, "A parallel radix 4 FFT computer,"IEEE Trans. Comput., vol. C-24, no. 1, pp. 80-92, Jan. 1975.
[17] M. J. Corinthios, Y. Geadah, and K. C. Smith, "Generalized spectral analysis algorithms and processors," inProc. Int. Conf. Inform. Sci. Syst., Patras, Greece, Aug. 19-24, 1976, pp. 257-262.

Index Terms:
pipeline processing; parallel architectures; matrix algebra; pipelined processing; matrices; parallel processing; general-base matrices; sampling matrices; computer architecture; Poles; zeros; pointers; spans; algorithm parameters; parallel pipelined processors; memory topology; access uniformity; shuffle complexity; algorithm factorizations; pipelined architecture; matrix theory; generalized perfect shuffle; Chrestenson generalized Walsh transform; generalized spectral analysis.
Citation:
M.J. Corinthios, "Optimal Parallel and Pipelined Processing Through a New Class of Matrices with Application to Generalized Spectral Analysis," IEEE Transactions on Computers, vol. 43, no. 4, pp. 443-459, April 1994, doi:10.1109/12.278482
Usage of this product signifies your acceptance of the Terms of Use.