Hardware Realization of Fuzzy Neural Networks

  • C. Hart Poskar
  • Peter J. Czezowski
  • Witold Pedrycz

Abstract

It is needless to say that a panoply of real-world applications of fuzzy sets call for a variety of systems realizing fuzzy computation. Concurrently, it is highly desirable to develop some universal computing modules that may be easily customized to meet required hardware — software specifications. This concept is not that new, as in two-valued logic we can easily encounter various categories of configurable and programmable devices (PLDs), ranging from PAL’s to FPGA’s. These devices are standardized general purpose logic units, which may be configured to perform specific functions. To follow a similar avenue, it is indispensable to identify a few generic processing modules that are complete enough, when considered from a functional point of view, for general computations with fuzzy linguistic variables. A family of logic-based neurons [1][2], emerges as a collection of processing operations whose role is to model logic-oriented processing dominant in the theory of fuzzy sets. These configurable architectures arising within this framework can directly cope with the topology of the problem at hand.

Keywords

Assure Convolution 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Pedrycz, W.,”Fuzzy neural networks and neurocomputations,” Fuzzy Sets and Systems, 56, pp. 1–28, 1993.Google Scholar
  2. [2]
    Pedrycz, W. and A.F. Rocha, “Fuzzy-set based models of neurons and knowledge-based networks,” IEEE Trans. on Fuzzy Systems, vol. 1, no. 5, pp. 254–266, 1993.CrossRefGoogle Scholar
  3. [3]
    Berenji, H.R. and P. Khedkar, “Learning and tuning fuzzy logic controllers through reinforcements,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 724–740, 1992.CrossRefGoogle Scholar
  4. [4]
    Eklund, P. and F. Klawonn, “Neural fuzzy logic programming,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 815–818, 1992.CrossRefGoogle Scholar
  5. [5]
    Horikawa, S., T. Furuhashi and Y. Uchikawa, “On fuzzy modeling using fuzzy neural networks with the back-propagation algorithm,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 801–806, 1992.CrossRefGoogle Scholar
  6. [6]
    Jang, J.R., “Self-learning fuzzy controllers based on temporal back propagation,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 714–723, 1992.CrossRefGoogle Scholar
  7. [7]
    Keller, J.M. and R. Krishnapuram, “Evidence aggregation networks for fuzzy logic inference,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 761–769, 1992.CrossRefGoogle Scholar
  8. [8]
    Simpson, P.K., “Fuzzy min-max neural networks — part 1: classification,” IEEE Trans. on Neural Networks, vol. 3, no. 5, pp. 776–786, 1992.CrossRefGoogle Scholar
  9. [9]
    Takagi, H. et al., “Neural networks designed on approximate reasoning architecture and their applications,” IEEE Trans. on Neural Networks, vol. 3 no. 5, pp. 752–760, 1992.MathSciNetCrossRefGoogle Scholar
  10. [10]
    Pedrycz, W., Fuzzy Sets Engineering, CRC Press, Boca Raton, FL, 1995.MATHGoogle Scholar
  11. [11]
    Dubois, D. and H. Prade, Possibility Theory — An Approach to Computerized Processing of Uncertainty, Plenum Press, New York, 1988.MATHGoogle Scholar
  12. [12]
    Pedrycz, W., C.H. Poskar, P.J. Czezowski, “A reconfigurable fuzzy neural network with in-situ learning”, IEEE Micro, vol. 15, no. 4, pp. 19–30, August 1995.CrossRefGoogle Scholar
  13. [13]
    Butnariu, D. and E.P. Klement, Triangular Norm-based Measures and Games with Fuzzy Coalitions, Kluwer Academic Publishers, Dordercht, 1993.MATHGoogle Scholar
  14. [14]
    Card, H. “Digital VLSI backpropagation networks”, Canadian J. Elect. & Comp. Eng., vol. 20, no. 1, pp. 15–23, 1995.Google Scholar
  15. [15]
    Clarckson, T.G., C.K. Ng and Y. Guan, “The pRAM: an adaptive VLSI chip,” IEEE Trans. on Neural Networks, vol. 4, no. 3, pp. 408–411, 1993.CrossRefGoogle Scholar
  16. [16]
    Dickson, J., R. McLeod and H. Card, “Stochastic arithmetic implementations of artificial neural networks with in-situ learning”, Proc. IEEE Int. Conf. Neural Networks, (San Francisco, CA), pp. 711-716, 1993.Google Scholar
  17. [17]
    Hamilton, A. et al, “Integrated pulse stream neural networks: results, issues, and pointers,” IEEE Trans. on Neural Networks, vol. 3, no. 3, pp. 385–393, 1992.CrossRefGoogle Scholar
  18. [18]
    Lehmann, C., M. Viredaz and F. Blayo, “A generic systolic array building block for neural networks with on-chip learning,” IEEE Trans. on Neural Networks, vol. 4, no. 3, pp. 400–407, 1993.CrossRefGoogle Scholar
  19. [19]
    Manduit, M. et al., “Lneuro 1.0: a piece of hardware LEGO for building neural network systems,” IEEE Trans. on Neural Networks, vol. 3, no. 3, pp. 414–421, 1992.CrossRefGoogle Scholar
  20. [20]
    Tomlinson Jr., M.S., DJ. Walker, and A. Sivilotti, “A digital neural network architecture for VLSI”, Proc. Int. Joint Conf. Neural Networks, (San Diego, CA), vol. 2, pp. 545–550, 1990.Google Scholar
  21. [21]
    Watanabe, T. et al, “A single 1.5-V digital chip for a 106 synapse neural network,” IEEE Trans. on Neural Networks, vol. 4, no. 3, pp. 387–393, 1993.CrossRefGoogle Scholar
  22. [22]
    Wawrzynek, J., K. Asanovic and N. Morgan, “The design of a neuro-microprocessor,” IEEE Trans. on Neural Networks, vol. 4, no. 3, pp. 394–399, 1993.CrossRefGoogle Scholar
  23. [23]
    Horst, R.W., “Task-flow architecture for WSI parallel processing,” Computer, vol. 25, no. 4, pp. 10–18, 1992.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1998

Authors and Affiliations

  • C. Hart Poskar
  • Peter J. Czezowski
  • Witold Pedrycz

There are no affiliations available

Personalised recommendations