ARX Model Identification using Generalized Spectral Decomposition

This article is concerned with the identification of autoregressive models exogenous inputs (ARX). Most existing approaches like prediction error minimization and state-space framework are widely accepted utilized to estimate ARX but known deliver unbiased consistent parameter estimates for a correctly supplied guess input-output orders delay. In this paper, we propose novel automated spectral decomposition that recovers orders, delay, output noise distribution, along estimates. The proposed algorithm systematically all parameters in two steps. first step order by examining generalized eigenvalues, second from eigenvectors. Simulation studies presented demonstrate method’s efficacy observed even at low signal-to-noise ratio (SNR).