Identification of Parametric Models in the Presence of Nonlinear Distortions
Johan Schoukens, Rik Pintelon, Yves Rolain · 2012
This chapter identifies the best linear approximation (BLA) of a nonlinear system. It then focuses on use of periodic excitation signals. The chapter uses the robust method to characterize the noise and the nonlinear distortion level. It identifies the best linear approximation starting from experiments with random excitations using time domain methods with parametric noise models, and frequency domain methods with nonparametric noise models. The studies the impact of nonlinear distortions on the uncertainty interval. Finally, the chapter identifies the best linear approximation starting from experiments with periodic excitations, using only one realization, or multiple realizations. Controlled Vocabulary Terms frequency response; predistortion