A semi-parametric polynomial frequency response estimator for identification of weakly non-linear systems using multi-amplitude MLS measurements - part II: Extension and application of the Hp estimator

In this article the family of polynomial estimators for the estimation of the linear system response of weakly non-linear systems is examined regarding estimation errors and applicability. The Hp3 estimator is introduced as an extension for measurement setups having both input and output noise and it is shown how the results obtained with the polynomial estimators are related to the parameters of Volterra models. The bias-variance tradeoff of the estimation error is derived and methods for obtaining the minimum error are presented.