Circuits, Systems, and Signal Processing

, Volume 38, Issue 12, pp 5755–5785 | Cite as

High-Performance System-on-Chip-Based Accelerator System for Polynomial Matrix Multiplications

  • Server KasapEmail author
  • Soydan Redif


Polynomial matrix computations, such as polynomial matrix multiplication (PMM) and eigenvalue factorization of parahermitian matrices, have played an important role in a growing number of applications, in recent times. However, the computational complexity and expense of such operations impose a profound limit on their applicability. In a recent paper, we introduced a systolic array-based parallel architecture for PMM, which was adequately efficient, but limited in its application. In this paper, we propose a second-generation hardware solution which boasts more versatility, efficiency and scalability compared to our previous design. This is achieved through the design of a highly versatile PMM accelerator which supports polynomial matrices of any size, as a component of the embedded system developed within the Xilinx Zynq-7000 AP SoC. Experimental results demonstrate the efficiency and effectiveness of our novel SoC-based PMM accelerator in the context of subband coding, where maximum speedups of \(85\times \) and \(33\times \) are accomplished, without compromising the accuracy, in comparison with two highly optimized and multi-threaded software-only implementations running on a dual-core ARM Cortex-A9 processor and a Intel Core i7-4510U CPU, respectively.


Polynomial matrix multiplication Polynomial matrix computations Computer architecture Hardware/software co-design System-on-chip (SoC) Zynq-7000 AP SoC 



  1. 1.
    M.A. Alrmah, S. Weiss, S. Lambotharan, An extension of the MUSIC algorithm to broadband scenarios using a polynomial eigenvalue decomposition. in Proceedings of European Signal Processing Conference, pp. 629–633 (2011)Google Scholar
  2. 2.
    R. Bracewell, The Fourier Transform and Its Applications (McGraw-Hill Higher Education, New York, 1999)zbMATHGoogle Scholar
  3. 3.
    R. Brandt, M. Bengtsson, Wideband MIMO channel diagonalization in the time domain. inProceedings of (IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, 2011), pp. 1958–1962Google Scholar
  4. 4.
    Digilent inc (2019) ZedBoard Zynq-7000 ARM/FPGA SoC Development Board. URL
  5. 5.
    J. Foster, J.G. McWhirter, S. Lambotharan, I. Proudler, M. Davies, J. Chambers, Polynomial matrix QR decomposition for the decoding of frequency selective multiple-input multiple-output communication channels. IET Signal Process. 6(7), 704–71 (2012)MathSciNetCrossRefGoogle Scholar
  6. 6.
    G.H. Golub, C.F.V. Loan, Matrix Computations (John Hopkins University Press, Baltimore, 1996)zbMATHGoogle Scholar
  7. 7.
    T. Kailath, Linear Systems (Prentice Hall, Upper Saddle River, 1980)zbMATHGoogle Scholar
  8. 8.
    S. Kasap, S. Redif, Novel field-programmable gate array architecture for computing the eigenvalue decomposition of para-Hermitian polynomial matrices. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 22(3), 522–536 (2014)CrossRefGoogle Scholar
  9. 9.
    V.K.P. Kumar, Y.C. Tsai, On synthesizing optimal family of linear systolic arrays for matrix multiplication. IEEE Trans. Comput. 40(6), 770–774 (1991)CrossRefGoogle Scholar
  10. 10.
    S.Y. Kung, Y. Wu, X. Zhang, Bezout space-time precoders and equalizers for MIMO channels. IEEE Trans. Signal Process. 50(10), 2499–2514 (2002)CrossRefGoogle Scholar
  11. 11.
    RH. Lambert, M. Joho, H. Mathis, Polynomial singular values for number of wideband source estimation and principal components analysis. in Proceedings of International Conference on Independent Component Analysis, pp. 379–383 (2001)Google Scholar
  12. 12.
    J.G. McWhirter, P.D. Baxter, T. Cooper, S. Redif, J. Foster, An EVD algorithm for para-Hermitian polynomial matrices. IEEE Trans. Signal Process. 55(5), 2158–2169 (2007)MathSciNetCrossRefGoogle Scholar
  13. 13.
    N. Moret, A. Tonello, S. Weiss, MIMO precoding for filter bank modulation systems based on PSVD. in Proceedings of IEEE Vehicular Technology Conference, pp. 1–5 (2011)Google Scholar
  14. 14.
    P. Moulin, M.K. Mihcak, Theory and design of signal-adapted FIR paraunitary filter banks. IEEE Trans. Signal Process. 46(4), 920–929 (1998)CrossRefGoogle Scholar
  15. 15.
    A.V. Oppenheim, C.J. Weinstein, Effects of finite register length in digital filtering and the fast Fourier transform. Proc. IEEE 60(8), 957–976 (1972)CrossRefGoogle Scholar
  16. 16.
    A. Papoulis, Probability, Random Variables, and Stochastic Processes (McGraw-Hill, New York, 1991)zbMATHGoogle Scholar
  17. 17.
    S. Redif, Fetal electrocardiogram estimation using polynomial eigenvalue decomposition. Turk. J. Electr. Eng. Comput. Sci. 24(4), 2483–2497 (2014)Google Scholar
  18. 18.
    S. Redif, Convolutive blind signal separation via polynomial matrix generalised eigenvalue decomposition. Electron. Lett. 53(2), 87–89 (2017)CrossRefGoogle Scholar
  19. 19.
    S. Redif, S. Kasap, Novel reconfigurable hardware architecture for polynomial matrix multiplications. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 23(3), 454–465 (2015)CrossRefGoogle Scholar
  20. 20.
    S. Redif, S. Weiss, J.G. McWhirter, An approximate polynomial matrix eigenvalue decomposition algorithm for para-Hermitian matrices. in Proceedings of IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), pp. 421–425 (2011)Google Scholar
  21. 21.
    S. Redif, S. Weiss, J.G. McWhirter, Sequential matrix diagonalization algorithms for polynomial EVD of Parahermitian matrices. IEEE Trans. Signal Process. 63(1), 81–89 (2015)MathSciNetCrossRefGoogle Scholar
  22. 22.
    P.A. Regalia, P. Loubaton, Rational subspace estimation using adaptive lossless filters. IEEE Trans. Signal Process. 40(10), 2392–2405 (1992)CrossRefGoogle Scholar
  23. 23.
    A. Tkacenko, Approximate eigenvalue decomposition of para-Hermitian systems through successive FIR paraunitary transformations. in Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 4074–4077 (2010)Google Scholar
  24. 24.
    P.P. Vaidyanathan, Multirate Systems and Filter Banks (Prentice Hall, Upper Saddle River, 1993)zbMATHGoogle Scholar
  25. 25.
    P.P. Vaidyanathan, Theory of optimal orthonormal subband coders. IEEE Trans. Signal Process. 46(6), 1528–1543 (1998)CrossRefGoogle Scholar
  26. 26.
    Z. Wang, J.G. McWhirter, S. Weiss, Multichannel spectral factorization algorithm using polynomial matrix eigenvalue decomposition. in Proceedings of Asilomar Conference on Signals, Systems and Computers, pp. 1714–1718 (2015)Google Scholar
  27. 27.
    S. Weiss, S. Redif, T. Cooper, C. Liu, P. Baxter, J.G. McWhirter, Paraunitary oversampled filter bank design for channel coding. EURASIP J. Appl. Signal Process. 2006, 1–10 (2006)zbMATHGoogle Scholar
  28. 28.
    S. Weiss, M. Alrmah, S. Lambotharan, J.G. McWhirter, M. Kaveh, Broadband angle of arrival estimation methods in a polynomial matrix decomposition framework. in Proceedings of IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, pp. 109–112 (2013)Google Scholar
  29. 29.
    S. Weiss, J. Pestana, I.K. Proudler, On the existence and uniqueness of the eigenvalue decomposition of a Parahermitian matrix. IEEE Trans. Signal Process. 66(10), 2659–2672 (2018)MathSciNetCrossRefGoogle Scholar
  30. 30.
  31. 31.
  32. 32.
    Xilinx inc (2015) FIFO Generator v12.0 LogiCORE IP Product Guide. URL
  33. 33.
    Xilinx inc (2017) Fast Fourier Transform v9.0 LogiCORE IP Product Guide. URL
  34. 34.
    Xilinx inc (2018a) 7 Series DSP48E1 Slice User Guide. URL
  35. 35.
    Xilinx inc (2018b) AXI DMA v7.1 LogiCORE IP Product Guide. URL
  36. 36.
    Xilinx inc (2018c) Zynq-7000 All Programmable SoC Data Sheet: Overview Data. URL
  37. 37.
    Xilinx inc (2018d) Zynq UltraScale+ MPSoC Data Sheet: Overview. URL
  38. 38.
    Xilinx inc (2019a) Vivado Design Suite. URL
  39. 39.
    Xilinx inc (2019b) Vivado System Generator for DSP. URL

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Computer Science and Electronic EngineeringUniversity of EssexEssexUK
  2. 2.Faculty of EngineeringEuropean University of LefkeLefkeTurkey

Personalised recommendations