论文信息 - Review and Unification of Linear Identifiability Concepts

Review and Unification of Linear Identifiability Concepts

To capture a unified perspective of the identification problem, we present a concise review of all contemporary concepts of identifiability. These concepts are axiomatically defined for a specific class of linear, deterministic and stochastic systems. From a derived set of criteria we demonstrate a basic correspondence between the principal concepts in discrete time. Such an equivalence statement constitutes the motif for the future development of a global identifiability concept applicable to nonlinear problems.

Eric F. Wood | V. V. Nguyen | V. Nguyen

[1] J. Aitchison,et al. Maximum-Likelihood Estimation of Parameters Subject to Restraints , 1958 .

[2] E. Tse. Information matrix and local identifiability of parameters , 1973 .

[3] L. Ljung. On Consistency and Identifiability , 1976 .

[4] Arthur Gelb,et al. Applied Optimal Estimation , 1974 .

[5] R. Kashyap. Maximum likelihood identification of stochastic linear systems , 1970 .

[6] E. Barankin. Sufficient parameters: Solution of the minimal dimensionality problem , 1960 .

[7] A. Jazwinski. Stochastic Processes and Filtering Theory , 1970 .

[8] Franklin M. Fisher,et al. Generalization of the Rank and Order Conditions for Identifiability , 1959 .

[9] B. Anderson,et al. Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[10] H. Akaike. Stochastic theory of minimal realization , 1974 .

[11] Graham C. Goodwin,et al. Coupled design of test signals, sampling intervals, and filters for system identification , 1974 .

[12] Regions of constrained maximum likelihood parameter identifiability , 1975, 1975 IEEE Conference on Decision and Control including the 14th Symposium on Adaptive Processes.