Continuous-time Zhang Neural Networks for AR Spectral Estimator

Abstract

In this paper, we present an Auto-Regressive (AR) spectral estimator using a special kind of recurrent neural network proposed by Zhang called Continuous-time Zhang Neural Network (CZNN) to solve a system of linear equations. This neural network is characterized by an implicit dynamics and designed by defining a vector-valued error function instead of the usual scalar-valued norm-based error function used in the Gradient based Neural Networks (GNN). The output of the CZNN is the estimated AR coefficients so that the spectrum of the signal can be directly obtained in terms of the AR coefficients. For comparative purposes, the GNN model is also employed for AR parameters estimation.