Resolve of Overlapped Voltammetric Peaks in Using a Wavelet Packet Transform Based Radial Basis Function Network

A novel method named WPTRBF is presented in this paper. This method is based on radial basis function neural network (RBFN) with wavelet packet transform (WPT) as a pre-processing tool for the simultaneous differential pulse stripping voltammetric determination of Pb(II), Ni(II) and Cd(II). Wavelet packet representations of signals provide a local time-frequency description, thus in the wavelet packet domain, the quality of noise removal can be improved. Radial basis function network was applied for overcoming the convergence problem met in back propagation training and for facilitating nonlinear calculation. In this case, through optimization, the number of wavelet function, decomposition level, the number of hidden nodes and the width (sigma) of RBFN for the WPTRBFN method were selected as Daubechies 4, 3, 8 and 0.7 respectively. The relative standard errors of prediction for all elements with WPTRBFN and RBFN were 5.8% and 6.89% respectively.