Advertisement

Second Order Function Approximation Using a Single Multiplication on FPGAs

  • Jérémie Detrey
  • Florent de Dinechin
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3203)

Abstract

This paper presents a new scheme for the hardware evaluation of elementary functions, based on a piecewise second order minimax approximation. The novelty is that this evaluation requires only one small rectangular multiplication. Therefore the resulting architecture combines a small table size, thanks to second-order evaluation, with a short critical path: Consisting of one table lookup, the rectangular multiplication, and one addition, the critical path is shorter than that of a plain first-order evaluation. Synthesis results for several functions show that this method outperforms all the previously published methods in both area and speed for precisions ranging from 12 to 24 bits and over.

Keywords

Critical Path Table Lookup Computer Arithmetic Input Word 15th IEEE Symposium 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Cao, J., Wei, B.W.Y., Cheng, J.: High-performance architectures for elementary function generation. In: Burgess, N., Ciminiera, L. (eds.) 15th IEEE Symposium on Computer Arithmetic, Vail, Colorado (June 2001)Google Scholar
  2. 2.
    Das Sarma, D., Matula, D.W.: Faithful bipartite ROM reciprocal tables. In: Knowles, S., McAllister, W.H. (eds.) 12th IEEE Symposium on Computer Arithmetic, Bath, UK, pp. 17–28. IEEE Computer Society Press, Los Alamitos (1995)Google Scholar
  3. 3.
    de Dinechin, F., Tisserand, A.: Some improvements on multipartite table methods. In: Burgess, N., Ciminiera, L. (eds.) 15th IEEE Symposium on Computer Arithmetic, Vail, Colorado, June 2001, pp. 128–135 (2001); Updated version of LIP research report 2000-38Google Scholar
  4. 4.
    Defour, D., de Dinechin, F., Muller, J.M.: A new scheme for table-based evaluation of functions. In: 36th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, California (November 2002)Google Scholar
  5. 5.
    Detrey, J., de Dinechin, F.: A VHDL library of LNS operators. In: 37th Asilomar Conference on Signals, Pacific Grove, USA (October 2003)Google Scholar
  6. 6.
    Hassler, H., Takagi, N.: Function evaluation by table look-up and addition. In: Knowles, S., McAllister, W.H. (eds.) 12th IEEE Symposium on Computer Arithmetic, Bath, UK, pp. 10–16. IEEE Computer Society Press, Los Alamitos (1995)Google Scholar
  7. 7.
    Lee, D.-U., Luk, W., Villasenor, J., Cheung, P.: Hierarchical segmentation schemes for function evaluation. In: IEEE Conference on Field-Programmable Technology, Tokyo (December 2003)Google Scholar
  8. 8.
    Lewis, D.M.: Interleaved memory function interpolators with application to an accurate LNS arithmetic unit. IEEE Transactions on Computers 43(8), 974–982 (1994)zbMATHCrossRefGoogle Scholar
  9. 9.
    Liddicoat, A.A.: High-performance arithmetic for division and the elementary functions. PhD thesis, Stanford University (2002)Google Scholar
  10. 10.
    Mencer, O., Boullis, N., Luk, W., Styles, H.: Parametrized function evaluation on fpgas. In: Field-Programmable Logic and Applications, Belfast (September 2001)Google Scholar
  11. 11.
    Muller, J.M.: Elementary Functions, Algorithms and Implementation. Birkhauser, Boston (1997)zbMATHGoogle Scholar
  12. 12.
    Muller, J.M.: A few results on table-based methods. Reliable Computing 5(3), 279–288 (1999)zbMATHCrossRefGoogle Scholar
  13. 13.
    Piñeiro, J.A., Bruguera, J.D., Muller, J.-M.: Faithful powering computation using table look-up and a fused accumulation tree. In: Burgess, N., Ciminiera, L. (eds.) 15th IEEE Symposium on Computer Arithmetic, Vail, Colorado, June 2001, pp. 40–47 (2001)Google Scholar
  14. 14.
    Stine, J.E., Schulte, M.J.: The symmetric table addition method for accurate function approximation. Journal of VLSI Signal Processing 21(2), 167–177 (1999)CrossRefGoogle Scholar
  15. 15.
    Vassiliadis, S., Zhang, M., Delgado-Frias, J.G.: Elementary function generators for neural-network emulators. IEEE transactions on neural networks 11(6), 1438–1449 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Jérémie Detrey
    • 1
  • Florent de Dinechin
    • 1
  1. 1.Laboratoire de l’Informatique du ParallélismeÉcole Normale Supérieure de LyonLyon cedex 07France

Personalised recommendations