A PARAMETERIZABLE HANDEL-C NEURAL NETWORK IMPLEMENTATION FOR FPGA

Cherrad Benbouchama, Mohamed Tadjine, Ahmed Bouridane

2009

Abstract

This paper shows the design possibility of a parameterizable implementation of neural multi-layer network on FPGA circuits (Field Programmable Gate Array) through the use of Handel-C language. The algorithm used for the training is the back-propagation. The tools of implementation and synthesis are the DK 4 of Celoxica and the ISE 6.3 of Xilinx. The targeted components are XCV2000 on Celoxica RC1000 board and XC2V1000 on RC200. The representation of the real numbers in fixed point was used for the data processing. The realization of the activation function is made with the approximate polynomial. A high level environment was designed in order to specify and introduce architecture parameters.

References

  1. Cox, C.E. and E. Blanz, GangLion, “ a fast field - programmable gate array implementation of a connectionist classifier ”, IEEE Journal of Solid-State Circuits, 1992. 28(3): p. 288-299.
  2. J. G. Eldredge and B. L. Hutchings, “ RRANN: A Hardware Implementation of the Backpropagation Algorithm Using Reconfigurable FPGAs ”, In IEEE World Conference on Computational Intelligence, Orlando, FL, 1994.
  3. A. T. Ferrucci, “ A Field-Programmable Gate Array Implementation of Self-Adapting and Scalable Connectionist Network ”, Mars 1994.
  4. M. Martin, “A reconfigurable hardware accelerator for back-propagation connectionist classifiers”, Masters thesis, University of California, Santa Cruz, 1994.
  5. H. Ossoinig, E. Reisinger, C. Steger, and R. Weiss, “Design and FPGA implementation of a neural network ”, In Proc. 7th Int. Conf. on Signal Processing Applications and Technology, pp. 939943, Orlando, Florida, 1996.
  6. J. Beuchat, J. Haenni, and E.Sanchez, “ Hardware Reconfigurable Neural Networks, In Parallel and Distributed Processing ”, IPPS/SPDP, pp. 91 98, Springer-Verlag, 1998.
  7. V. Pandya, “ A Handel-C implementation of thebackpropagation algorithm on field programmable gate arrays ”, Master thesis. Faculty of Graduate Studies of the University of Guelph, Canada, December 2005.
  8. C. Benbouchama and al, “ The FPGA Neural Networks Implementation for a Real Time Control ”, Archives of Control Sciences (A.C.S.), Vol. 17(LIII), n. 1, pp. 527, 2007.
  9. T. Stöcklein and J. Bhig, “ Handel-C an effective method for designing FPGAs (and ASICs) ”, Georg SimonOhm Fachhochschule, NÜRNBERG 2002.
  10. P. J. Ashenden, “ The Designer's Guide To VHDL ”, second edition, San Diego: Morgan Kanfmann, 2002.
  11. J. Hopf, “ Comparing the Bitstreams of Applications Specified in Hardware Join Java and HandelC ”, 2nd IEEE International Conference on Field
  12. Programmable Technology, Tokyo, Japan, 2003. Celoxica: www.Celoxica.com
Download


Paper Citation


in Harvard Style

Benbouchama C., Tadjine M. and Bouridane A. (2009). A PARAMETERIZABLE HANDEL-C NEURAL NETWORK IMPLEMENTATION FOR FPGA . In Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO, ISBN 978-989-8111-99-9, pages 325-328. DOI: 10.5220/0002166903250328


in Bibtex Style

@conference{icinco09,
author={Cherrad Benbouchama and Mohamed Tadjine and Ahmed Bouridane},
title={A PARAMETERIZABLE HANDEL-C NEURAL NETWORK IMPLEMENTATION FOR FPGA},
booktitle={Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,},
year={2009},
pages={325-328},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002166903250328},
isbn={978-989-8111-99-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,
TI - A PARAMETERIZABLE HANDEL-C NEURAL NETWORK IMPLEMENTATION FOR FPGA
SN - 978-989-8111-99-9
AU - Benbouchama C.
AU - Tadjine M.
AU - Bouridane A.
PY - 2009
SP - 325
EP - 328
DO - 10.5220/0002166903250328