FPLD HDL synthesis employing high-level evolutionary algorithm optimisation | SpringerLink
Skip to main content

FPLD HDL synthesis employing high-level evolutionary algorithm optimisation

  • Design Tools
  • Conference paper
  • First Online:
Field-Programmable Logic and Applications (FPL 1997)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 1304))

Included in the following conference series:

Abstract

This paper presents a novel approach to optimising high level designs for circuits to be implemented on FPLDs. The aim is to search the design space using an evolutionary algorithm to find solutions that optimise circuit speed and circuit size under given constraints. To ensure correct circuit operation, a library of synchronous functional modules are used and interchanged in the circuit, altering only each module's data type (not its functionality). After modifying a circuit, modules for data synchronising and type conversions are added automatically. It is these extra modules that cause the search to become non-linear, indicating that the combination of optimised sub-circuits does not necessarily give an overall optimised circuit. The input to the synthesiser and optimiser is a netlist of modules, while the output is a completely specified Altera Hardware Description Language (AHDL) listing ready to be compiled. The main advantages of the method are that existing sub-circuits can be utilised, circuits can often be fit into available hardware without being redesigned, advances in algorithms and sub-circuit designs can be utilised, and low-level compilers and optimisers are left to their speciality.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Salcic, Z., and Maunder, B.: CCSimP-An Instruction-Level Custom-Configurable Processor for FPLDs. Proceedings 6th International Workshop on Field-Programmable Logic and Applications, Darmstadt, Germany, pages 280-289, September 1996

    Google Scholar 

  2. van Daalen, M., Jeavons, P., and Shawe-Taylor, J.: A Stochastic Neural Architecture That Exploits Dynamically Reconfigurable FPGAs. Proceedings IEEE Workshop on FPGAs for Custom Computing Machines, pages 202-211, April 1993.

    Google Scholar 

  3. Bade, S.L., and Hutchings, B.L.: FPGA-Based Stochastic Neural Networks-Implementation. In Proceedings of IEEE Workshop on FPGAs for Custom Computing Machines, Napa, CA, pages 189-198, April 1994.

    Google Scholar 

  4. van Daalen, M., Jeavons, P., Shawe-Taylor, J., and Cohen, D.: Device For Generating Binary Sequences for Stochastic Computing. Electronic Letters, 29(1):80–81, Jan 1993.

    Google Scholar 

  5. Battersby, A.: Network Analysis for Planning and' Scheduling: 2nd Ed. MacMillan and Company Ltd., London, 1967.

    Google Scholar 

  6. Goldberg, D.E.: Genetic Algorithms: in Search, Optimization and Machine Learning. Addison-Wesley, 1989.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Wayne Luk Peter Y. K. Cheung Manfred Glesner

Rights and permissions

Reprints and permissions

Copyright information

© 1997 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Maunder, R., Salcic, Z.A., Coghill, G.G. (1997). FPLD HDL synthesis employing high-level evolutionary algorithm optimisation. In: Luk, W., Cheung, P.Y.K., Glesner, M. (eds) Field-Programmable Logic and Applications. FPL 1997. Lecture Notes in Computer Science, vol 1304. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63465-7_231

Download citation

  • DOI: https://doi.org/10.1007/3-540-63465-7_231

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-63465-2

  • Online ISBN: 978-3-540-69557-8

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics