Intelligent Signal Estimation Using Cosine Neural Networks with Variable Structure Systems Based Training Procedure

Abstract

This paper demonstrates the estimation of signals by using a neural network structure composed of cosine neurons. The building blocks of the architecture are cosine components with adjustable amplitude, frequency and phase. The training procedure is based on the mixture of gradient descent with a method utilizing sliding mode control philosophy. The… (More)

9 Figures and Tables

Topics

  • Presentations referencing similar topics