Application of the narmax method to the modelling of the nonlinearity of dynamic loudspeakers

Downloads

Authors

  • A. B. DOBRUCKI Institute of Telecommunications and Acoustics, Wrocław University of Technology
  • P. PRUCHNICKI Institute of Telecommunications and Acoustics, Wrocław University of Technology

Abstract

The application of the NARMAX method to the modelling of the nonlinearity of dynamic loudspeakers is described. The principle of creating a polynomial representation of a model, the problems stemming from a too large number of model coefficients and the method of optimizing the model are presented. The method was tested on data from actual loudspeaker measurements. Different models are compared as regards their accuracy depending on the modelling parameters. Finally, the model characteristics are compared with the results of loudspeaker measurements performed by classical methods.

References

[1] S.A. Billings and S. Chen, Extended model set, global data and threshold model identification of severely non-linear systems, Int. J. of Control, 50, 1897-1923 (1989).

[2] S.A. Billings and S. Chen, Identification of non-linear rational systems using a prediction-error estimation algorithm, Int. J. of Systems Sci., 20, 467-494 (1989).

[3] S.A. Billings and W.S.F. Voon, Least squares parameter estimation algorithms for non-linear systems, Int. J. of Systems Sci., 15, 601-615 (1984).

[4] S. Chen and S.A. Billings, Representations of non-linear systems: the NARMAX model, Int. J. of Control, 89, 1013-1032 (1989).