1 May 1991 in IEEE Transactions on Automatic Control
The traditional approach of expanding transfer functions and noise models in the delay operator to obtain linear-in-the-parameters predictor models leads to approximations of very high order in cases of rapid sampling and/or dispersion in time constants. By using prior information about the time constants of the system more appropriate expansions, related to Laguerre networks, are introduced and analyzed. It is shown that the model order can be reduced, compared to ARX (FIR, AR) modeling, by using Laguerre models. Furthermore, the numerical accuracy of the corresponding linear regression estimation problem is improved by a suitable choice of the Laguerre parameter. Consistency (error bounds), persistence of excitation conditions. and asymptotic statistical properties are investigated. This analysis is based on the result that the covariance matrix of the regression vector of a Laguerre model has a Toeplitz structure.<>
1 June 1994 in IEEE Transactions on Automatic Control
1 July 2012 in IFAC Proceedings Volumes
1 July 1992 in IEEE Transactions on Automatic Control
1 February 1986 in IEEE Transactions on Automatic Control
Estimation of transfer functions of linear systems is one of the most common system identification problems. Several different design variables, chosen by the user for the identification procedure, affect the properties of the resulting estimate. In this paper it is investigated how the choices of prefilters, noise models, sampling interval, and prediction horizon (i.e., the use ofk-step ahead prediction methods) influence the estimate. An important aspect is thai the true system is not assumed to be exactly represented within the chosen model set. The estimate will thus be biased. It is shown how the distribution of bias in the frequency domain is governed by a weighting function, which emphasizes different frequency bands. The weighting function, in turn, is a result of the previously listed design variables. It is shown, e.g., thai the common least-squares method has a tendency to emphasize high frequencies, and that this can be counteracted by prefiltering. It is also shown that, asymptotically, it is only the prediction horizon itself, and not how it is split up into sampling interval times number of predicted sampling instants, that affects this weighting function.