Numerically robust transfer function modeling from noisy frequency domain data

Using vector orthogonal polynomials as basis functions for the representation of the rational form of a linear time invariant system, in frequency domain identification problems, it is shown that the notorious numerical ill conditioning of these maximum likelihood problems can be overcome completely. For the identification of high-order (100/100) systems operating over a wide frequency band, or even in the situation of over- or undermodeling, condition numbers less than ten are reported for real measurements.

[1]  Adhemar Bultheel,et al.  A parallel algorithm for discrete least squares rational approximation , 1992 .

[2]  Y. Rolain.  Generating Robust Starting Values for Frequency Domain Transfer Function Estimation. , 1996 .

[3]  G. Golub,et al.  Updating and downdating of orthogonal polynomials with data fitting applications , 1991 .

[4]  B. Ninness,et al.  A unifying construction of orthonormal bases for system identification , 1997, IEEE Trans. Autom. Control..

[5]  Gerd Vandersteen,et al.  Frequency-domain system identification using non-parametric noise models estimated from a small number of data sets , 1997, Autom..

[6]  R. Pintelon,et al.  On the use of orthogonal polynomials in high order frequency domain system identification and its application to modal parameter estimation , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[7]  Thomas Kailath,et al.  Linear Systems , 1980 .

[8]  Gerd Vandersteen,et al.  Model selection through a statistical analysis of the global minimum of a weighted nonlinear least squares cost function , 1997, IEEE Trans. Signal Process..

[9]  Yves Rolain,et al.  Generating robust starting values for frequency-domain transfer function estimation , 1999, Autom..

[10]  Adhemar Bultheel,et al.  Discrete linearized least-squares rational approximation on the unit circle , 1993 .

[11]  E. C. Levy.  Complex-curve fitting , 1959, IRE Transactions on Automatic Control.

[12]  J. Schoukens,et al.  Parametric and nonparametric identification of linear systems in the presence of nonlinear distortions-a frequency domain approach , 1998, IEEE Trans. Autom. Control..

[13]  J. Schoukens,et al.  Model selection through a statistical analysis of the global minimum of a weighted non-linear least squares cost function , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[14]  Adhemar Bultheel,et al.  Vector Orthogonal Polynomials and Least Squares Approximation , 1995, SIAM J. Matrix Anal. Appl..

[15]  William B. Gragg,et al.  Downdating of Szego Polynomials and Data Fitting Applications , 1992 .

[16]  Rik Pintelon,et al.  On the frequency scaling in continuous-time modeling , 2005, IEEE Transactions on Instrumentation and Measurement.

[17]  Joe Brewer,et al.  Kronecker products and matrix calculus in system theory , 1978 .

[18]  Adhemar Bultheel,et al.  Orthonormal polynomial vectors and least squares approximation for a discrete inner product. , 1995 .

[19]  Yves Rolain,et al.  Analyses, development and applications of TLS algorithms in frequency domain system identification , 1997 .

[20]  J. D. Ullman,et al.  Best Conditioned Parametric Identification of Transfer Function Models in the Frequency Domain , 1995 .

[21]  J. Schoukens,et al.  Parametric identification of transfer functions in the frequency domain-a survey , 1994, IEEE Trans. Autom. Control..

[22]  C. Sanathanan,et al.  Transfer function synthesis as a ratio of two complex polynomials , 1963 .