Fast Fixed-Point Optimization of DSP Algorithms

  • Gabriel Caffarena
  • Ángel Fernández-Herrero
  • Juan A. López
  • Carlos Carreras
Conference paper
Part of the IFIP Advances in Information and Communication Technology book series (IFIPAICT, volume 373)

Abstract

In this chapter, the fast fixed-point optimization of Digital Signal Processing (DSP) algorithms is addressed. A fast quantization noise estimator is presented. The estimator enables a significant reduction in the computation time required to perform complex fixed-point optimizations, while providing a high accuracy. Also, a methodology to perform fixed-point optimization is developed.

Affine Arithmetic (AA) is used to provide a fast Signal-to-Quantization Noise-Ratio (SQNR) estimation that can be used during the fixed-point optimization stage. The fast estimator covers differentiable non-linear algorithms with and without feedbacks. The estimation is based on the parameterization of the statistical properties of the noise at the output of fixed-point algorithms. This parameterization allows relating the fixed-point formats of the signals to the output noise distribution by means of fast matrix operations. Thus, a fast estimation is achieved and the computation time of the fixed-point optimization process is significantly reduced.

The proposed estimator and the fixed-point optimization methodology are tested using a subset of non-linear algorithms, such as vector operations, IIR filter for mean power computation, adaptive filters – for both linear and non-linear system identification – and a channel equalizer. The computation time of fixed-point optimization is boosted by three orders of magnitude while keeping the average estimation error down to 6% in most cases.

Keywords

Fixed-Point Optimization Digital Signal Processing Quantization Word-Length Affine Arithmetic Error Estimation Signal-to-Quantization-Noise Ratio 

References

  1. 1.
    Sung, W., Kum, K.-I.: Simulation-Based Word-Length Optimization Method for Fixed-Point Digital Signal Processing Systems. IEEE Trans. Signal Processing 43(12), 3087–3090 (1995)CrossRefGoogle Scholar
  2. 2.
    Constantinides, G.A., Cheung, P.Y.K., Luk, W.: Wordlength Optimization for Linear Digital Signal Processing. IEEE Trans. Computer-Aided Design 22(10), 1432–1442 (2003)CrossRefGoogle Scholar
  3. 3.
    Caffarena, G., Constantinides, G., Cheung, P., Carreras, C., Nieto-Taladriz, O.: Optimal Combined Word-Length Allocation and Architectural Synthesis of Digital Signal Processing Circuits. IEEE Trans. Circuits Syst. II 53(5), 339–343 (2006)CrossRefGoogle Scholar
  4. 4.
    Constantinides, G., Woeginger, G.: The Complexity of Multiple Wordlength Assignment. Applied Mathematics Letters 15, 137–140 (2002)MathSciNetMATHCrossRefGoogle Scholar
  5. 5.
    López, J., Caffarena, G., Carreras, C., Nieto-Taladriz, O.: Fast and accurate computation of the roundoff noise of linear time-invariant systems. IET Circuits, Devices & Systems 2(4), 393–408 (2008)CrossRefGoogle Scholar
  6. 6.
    Menard, D., Sentieys, O.: A Methodology for Evaluating the Precision of Fixed-Point Systems. In: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 3, pp. 3152–3155 (2002)Google Scholar
  7. 7.
    Constantinides, G.: Perturbation Analysis for Word-Length Optimization. In: Proc. FCCM, pp. 81–90 (2003)Google Scholar
  8. 8.
    Menard, D., Rocher, R., Scalart, P., Sentieys, O.: SQNR Determination in Non-Linear and Non-Recursive Fixed-Point Systems. In: Proc. EUSIPCO, pp. 1349–1352 (2004)Google Scholar
  9. 9.
    Shi, C., Brodersen, R.: A Perturbation Theory on Statistical Quantization Effects in Fixed-Point DSP with Non-Stationary Inputs. In: Proc. ISCAS, vol. 3, pp. 373–376 (2004)Google Scholar
  10. 10.
    Lee, D.-U., Gaffar, A., Cheung, R., Mencer, W., Luk, O., Constantinides, G.: Accuracy-Guaranteed Bit-Width Optimization. IEEE Trans. Computer-Aided Design 25(10), 1990–2000 (2006)CrossRefGoogle Scholar
  11. 11.
    Constantinides, G., Cheung, P., Luk, W.: Truncation Noise in Fixed-Point SFGs. IEE Electronics Letters 35(23), 2012–2014 (1999)CrossRefGoogle Scholar
  12. 12.
    Jackson, L.: Roundoff-noise analysis for fixed-point digital filters realized in cascade or parallel form. IEEE Trans. Audio Electroacoust. 18, 107–122 (1970)CrossRefGoogle Scholar
  13. 13.
    Rocher, R., Menard, D., Sentieys, O., Scalart, P.: Analytical Accuracy Evaluation of Fixed-Point Systems. In: Proc. EUSIPCO, pp. 999–1003 (2007)Google Scholar
  14. 14.
    Stolfi, J., Figueiredo, L.H.: Self-Validated Numerical Methods and Applications. In: Brazilian Mathematics Colloquium: IMPA (1997)Google Scholar
  15. 15.
    Hayes, B.: A Lucid Interval. American Scientist 91(6), 484–488 (2003)Google Scholar
  16. 16.
    López, J.: Evaluación de los Efectos de Cuantificación en las Estructuras de Filtros Digitales Mediante Técnicas de Simulación Basadas en Extensiones de Intervalos. PhD thesis, Universidad Politécnica de Madrid (2004)Google Scholar
  17. 17.
    López, J., Carreras, C., Nieto-Taladriz, O.: Improved Interval-Based Characterization of Fixed-Point LTI Systems With Feedback Loops. IEEE Trans. Computer-Aided Design 26, 1923–1933 (2007)CrossRefGoogle Scholar
  18. 18.
    Fang, C., Chen, T., Rutenbar, R.: Floating-Point Error Analysis Based on Affine Arithmetic. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, vol. 2, pp. 561–564 (2003)Google Scholar
  19. 19.
    López, J., Caffarena, G., Carreras, C., Nieto-Taladriz, O.: Analysis of Limit Cycles by Means of Affine Arithmetic Computer-aided Tests. In: Proc. EUSIPCO, pp. 991–994 (2004)Google Scholar
  20. 20.
    Fang, C., Rutenbar, R., Chen, T.: Fast, accurate static analysis for fixed-point finite-precision effects in DSP designs. In: Int. Conf. on Computer Aided Design, pp. 275–282 (November 2003)Google Scholar
  21. 21.
    Caffarena, G.: Combined Word-Length Allocation and High-Level Synthesis of Digital Signal Processing Circuits. PhD thesis, Universidad Politécnica de Madrid (2008)Google Scholar
  22. 22.
    Choi, H., Burleson, W.: Search-based Wordlength Optimization for VLSI/DSP Synthesis. In: IEEE Workshop VLSI Signal Processing, pp. 198–207 (1994)Google Scholar
  23. 23.
    Caffarena, G., Carreras, C.: Architectural synthesis of DSP circuits under simultaneous error and time constraints. In: Proc. IEEE/IFIP Int. Conference VLSI-SOC, pp. 322–327 (September 2010)Google Scholar
  24. 24.
    Cantin, M.-A., Savaria, Y., Prodanos, D., Lavoie, P.: An Automatic Word Length Determination Method. In: IEEE Int. Symp. on Circuits and Systems, Sydney, Australia, vol. 5, pp. 53–56 (2001)Google Scholar
  25. 25.
    Han, K., Evans, B., Swartzlander, E.: Data Wordlength Reduction for Low-Power Signal Processing Software. In: IEEE Workshop on Signal Processing Systems, pp. 343–348 (2004)Google Scholar
  26. 26.
    Catthoor, F., Vandewalle, J., De Man, H.: Simulated Annealing Based Optimization of Coefficient and Data Word-Lengths in Digital Filters. J. Circuit Theory Applications 16, 371–390 (1988)CrossRefGoogle Scholar
  27. 27.
    Todman, T., Constantinides, G., Wilton, S., Mencer, O., Luk, W., Cheung, P.: Reconfigurable computing: architectures and design methods. IEE Proceedings Computers and Digital Techniques 152, 193–207 (2005)CrossRefGoogle Scholar
  28. 28.
    Han, K., Evans, B.: Optimum Wordlength Search Using Sensivity Information. EURASIP Journal of Applied Signal Processing 2006, 1–14 (2006), doi:10.1155/ASP/2006/92849Google Scholar
  29. 29.
    Fernández, A., Jimenez, A., Caffarena, G., Casajús, J.: Design and Implementation of a Hardware Module for Equalisation in a 4G MIMO Receiver. In: Proc. FPL, pp. 765–768 (2006)Google Scholar
  30. 30.
    Haykin, S.: Adaptive Filter Theory. Prentice-Hall, Upper Saddle River (2002)Google Scholar
  31. 31.
    Ogunfunmi, T.: Adaptive Nonlinear System Identification: The Volterra and Wiener Approaches. Springer, Heidelberg (2007)MATHCrossRefGoogle Scholar

Copyright information

© IFIP International Federation for Information Processing 2012

Authors and Affiliations

  • Gabriel Caffarena
    • 1
  • Ángel Fernández-Herrero
    • 2
  • Juan A. López
    • 2
  • Carlos Carreras
    • 2
  1. 1.Dep. Ingeniería de Sistemas de Información y TelecomunicaciónUniversidad CEU San PabloMadridSpain
  2. 2.Dep. Ingeniería ElectrónicaUniversidad Politécnica de MadridMadridSpain

Personalised recommendations