论文信息 - Study of conditional ML estimators in time and frequency domain system identification

Study of conditional ML estimators in time and frequency domain system identification

This paper studies the impact of replacing the full covariance matrix by its main diagonal in maximum likelihood estimation (MLE) of linear dynamic systems in the time or frequency domain. First the source of the nondiagonal entries is studied, next the impact of neglecting these terms on the efficiency of the estimators is analyzed. Finally the equivalence between the time and frequency domain formulation is shown.

[1] T. Söderström. Convergence properties of the generalised least squares identitication method , 1974, Autom..

[2] C. K. Yuen,et al. Theory and Application of Digital Signal Processing , 1978, IEEE Transactions on Systems, Man, and Cybernetics.

[3] Gene H. Golub,et al. Matrix computations , 1983 .

[4] J. Schoukens,et al. Modeling the noise influence on the Fourier coefficients after a discrete Fourier transform , 1986, IEEE Transactions on Instrumentation and Measurement.

[5] Lennart Ljung,et al. System Identification: Theory for the User , 1987 .

[6] Petre Stoica,et al. Decentralized Control , 2018, The Control Systems Handbook.

[7] Hugo Van hamme,et al. Identification of linear dynamic systems using piecewise constant excitations: Use, misuse and alternatives , 1994, Autom..

[8] Gerd Vandersteen,et al. General framework for asymptotic properties of generalized weighted nonlinear least-squares estimators with deterministic and stochastic weighting , 1996, IEEE Trans. Autom. Control..

[9] Yves Rolain,et al. Maximum likelihood estimation of errors-in-variables models using a sample covariance matrix obtained from small data sets , 1997 .

[10] David R. Brillinger,et al. Time Series , 2018, Randomization, Bootstrap and Monte Carlo Methods in Biology.