Mlp neural network and on-line backpropagation learning implementation in a low-cost fpga

ABSTRACT

This paper presents an implementation of a multilayer perceptronneural network and the backpropagation learning algorithm in an FPGA. The resulting implementation, in contrast to others, is a low-cost system with effective resource utilization, capable of training the neural network for any given task. The system is based on a modular scheme conforming to a system-on-a-chip (SoC), where modules can be replaced or scaled to suit a specific application. The system uses fixed-point arithmetic and it was carried out using generic hardware description language. A pipeline architecture is used in order to build a time-efficient system. The efficacy of the systems was tested in a pattern recognition application, tests were done in a low-cost Xilinx Spartan-3E FPGA.

References

1. R. Castañeda-Miranda, E. Ventura-Ramos, R. del R. Peniche-Vera,and G. Herrera-Ruiz. Fuzzy greenhouse climate control system based on a field programmable gate array. Biosystems Engineering, 94(2):165--177, 2006.
2. R. de J. Romero-Troncoso. Sistemas Digitales con VHDL, chapter 1, pages 1--8. Legaria Ediciones, 2004.
3. R. de J. Romero-Troncoso, G. Herrera-Ruiz, I. Terol-Villalobos,and J. Jauregui-Correa. Fpga based on-line tool breakage detection system for cnc milling machines. Mechatronics, 14:439--454, 2004.
4. P. O. Domingos, F. M. Silva, and H. C. Neto. An efficient and scalable architecture for neural networks with backpropagation learning. International Conference on Field Programmable Logic and Applications, pages 89--94, 2005.
5. R. Gadea, J. Cerda, F. Ballester, and A. Macholi. Artificial neural network implementation on a single fpga of a pipelined on-line backpropagation. Proceedings of the 13th International Symposium on System Synthesis, pages 225--230, 2000.
6. S. Himavathi, D. Anitha, and A. Muthuramalingam. Feedforward neural network implementation in fpga using layer multiplexing for effective resource utilization. IEEE Trans. on Neural Networks, 18(3):880--888, May 2007.
7. S.-S. Kim and S. Jung. Hardware implementation of a neural network controller with an mcu and an fpga for nonlinear systems. International Journal of Control,Automation and Systems, 4(5):567--574, October 2006.
8. M. Lundh, F. Lundell, S. Zahrai, and D. Söderberg. Field programmable gate arrays for flexible and fast data processing. FPGA and structured ASIC Journal, March 2006.
9. J. Mendoza-Jasso, G. Ornelas-Vargas, R. C. Miranda, E. Ventura-Ramos, A. Zepeda-Garrido, and G. H. Ruiz. Fpga-based real-time remote monitoring system. Computer and Electronics in Agriculture, 49(2):272--285, 2005.
10. A. Muthuramalingam, S. Himavathi, and E. Srinivasan. Neural network implementation using fpga: Issues and application. International Journal of Information Technology, 4(2):86--92,2007.
11. A. W. Savich, M. Moussa, and S. Areibi. The impact of arithmetic representation on implementing mlp-bp on fpgas: A study. IEEE Trans. on Neural Networks, 18(1):240--252, January 2007.
12. B. Schrauwen and M. D. Haene. Compact digital hardware implementations of spiking neural networks. Sixth FirW PhD Symposium, page pp. on CD, 2005.
13. F. Smach, M. Atri, J. Mitéran, and M. Abid. Design of a neural network classifier for face detection. Journal of Computer Science, 2(3):257--260, 2006.

Index Terms

• Hardware

  • Very large scale integration design

    • Application-specific VLSI designs

      • Application specific processors