A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
A resource-allocating network for function interpolation
Neural Computation
Ten lectures on wavelets
Neural Networks: A Comprehensive Foundation
Neural Networks: A Comprehensive Foundation
Foundations of Wavelet Networks and Applications
Foundations of Wavelet Networks and Applications
Output value-based initialization for radial basis function neural networks
Neural Processing Letters
Wavelet neural networks for function learning
IEEE Transactions on Signal Processing
| Hi-index | 0.00 |
In this paper we focus on wavelet neural network (WNN) for approximating non linear functions with B-spline orthonormal scaling function as activation function. The orthonormal scaling functions allow significant reduction of computational complexity and results in a compact network structure. The system of activation function is linearly independent by definition and has the advantage of numerical stability. A learning procedure for the proposed WNN with guaranteed convergence to the global minimum error in the parameter function space is developed. The approximation capabilities are illustrated through experimentations. The proposed network has advantages of approximation accuracy and good generalization performance. The simulation results indicate the efficiency of the proposed approach.