Rational Model Identification Using an Extended Least Squares Algorithm

摘要

A new least squares based parameter estimation algorithm is derived for nonlinearsystems which can be represented by a rational model defined as the ratio of two polynomial expansions of past system inputs, outputs and noise. This new algorithm overcomes previous noise problems by removing the bias in both the normal matrix and the correlation vector for the rational model. The algorithm is much simpler than nonlinear least square procedure because it avoids complex minimization routines. Simulation results are included to illustrate the performance of the new algorithm.