论文信息 - Parameter estimation in linear dynamic systems

Parameter estimation in linear dynamic systems

We give some fairly general circumstances under which the experimental design problem for linear dynamic systems with correlated noise can be put in the exact form of the Kiefer-Wolfowitz experimental design problem. Optimal inputs are characterized in terms of their spectral measures with respect to a resolution of the identity in a certain reproducing kernel Hilbert space. Numerical methods established for the Kiefer-Wolfowitz design problem then apply to the numerical determination of an optimal spectral measure.

Grace Wahba | G. Wahba

[1] E. Parzen. An Approach to Time Series Analysis , 1961 .

[2] W. J. Studden,et al. Optimum designs when the observations are second-order processes , 1978 .

[3] W. J. Studden,et al. A Kiefer-Wolfowitz Theorem in a Stochastic Process Setting , 1979 .

[4] G. Goodwin,et al. Frequency domain approach for designing sampling rates for system identification , 1975, Autom..

[5] Changbao Wu,et al. Some iterative procedures for generating nonsingular optimal designs , 1978 .

[6] L. Keviczky. "Design of Experiments" for the Identification of Linear Dynamic Systems , 1975 .

[7] Andrej Pázman. Hilbert-space methods in experimental design , 1978, Kybernetika.

[8] D. Chang. Design of Optimal Control for a Regression Problem , 1979 .

[9] Changbao Wu,et al. Some Algorithmic Aspects of the Theory of Optimal Designs , 1978 .

[10] Raman K. Mehra,et al. Optimal input signals for parameter estimation in dynamic systems--Survey and new results , 1974 .