Application of adaptive RBF networks to odour classification using conducting polymer sensor array

Abstract

This paper presents the application of adaptive RBF(Radial Basis Function) Networks to odour classification using conducting polymer sensor array. The performance of RBF Network is highly dependent on the choice of centers and widths in basis function. For the fine tuning of centers and widths, those are initialized by ill-conditioned genetic fuzzy c-means algorithm and the distribution of input patterns in the very first leaning stage, the stochastic gradient (SG) method is adapted. The adaptive RBF Network, which has tuned centers and width by SG method, has shown good classification performance for complex and mixture chemical patterns and confirmed by experimental trails.