A Reconfigurable Fuzzy Neural Network with In-Situ Learning

  • Authors:

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

  • Affiliations:

  • -;-;-

  • Venue:
  • IEEE Micro
  • Year:
  • 1995

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a Reconfigurable Fuzzy Processor (RFP), for implementing both aggregative (OR, AND) and referential (Matching, Difference, Dominance, and Inclusion) operations. The RFP architecture combines both structural and parametric flexibility in the design of a network implementing these processors as a collection of fuzzy neurons. RFP neurons are composed of three main units: the fuzzy processor (also termed RFP), the learning unit (incorporating a Fuzzy Backpropagation learning algorithm) and a local memory unit.A Fuzzy Neural Network (FNN) is implemented as a bidirectionally linked series of shared buses. This implementation facilitates a modular and scalable design environment for the RFP. An appropriate interface is also provided, separate from the RFP neuron itself, to promote the reuse of the neuron design with alternative interconnect networks.