Highly Efficient Identification Methods for Dual-Rate Hammerstein Systems

Abstract

This brief concerns parameter identification for a dual-rate Hammerstein CARMA system. By combining the polynomial transformation technique and the hierarchical identification principle, this brief transforms a dual-rate nonlinear Hammerstein CARMA system into a bilinear dual-rate identification model, and presents a hierarchical least squares algorithm to estimate the parameter vectors of the bilinear dual-rate identification model. Moreover, by using the key term separation principle, this brief transforms the dual-rate nonlinear Hammerstein CARMA system into a linear dual-rate identification model, and presents a key term separation based least squares algorithm to estimate the parameter vector of the linear dual-rate identification model. The two proposed methods possess higher computational efficiency compared with the previous over-parameterization least squares method in which many redundant parameters need estimating. The simulation results show the effectiveness of the two proposed algorithms.

A novel data filtering based multi-innovation stochastic gradient algorithm for Hammerstein nonlinear systems

“…In the simulation examples, the proposed F-MISG algorithm is more accurate than the MI-ESG algorithm for the same innovation length and the introduction of the forgetting factor can effectively improve the parameter estimation accuracy of the multi-innovation identification algorithms. The basic idea of the proposed methods can be extended to identify other linear systems [38,39] and nonlinear systems [40][41][42] with colored noise.…”

The model equivalence based parameter estimation methods for Box–Jenkins systems

“…The simulation results indicate that the proposed algorithm has higher accuracy than the auxiliary model based recursive generalized extended least squares algorithm. The proposed method can be extended to study identification problems of other dual-rate multivariable linear systems with colored noise [35,36,37], nonlinear systems with colored noise [38,39,40] and so on [41,42]. …”

“…Comparing the information vector in (36) with that in (18), it can be seen that the noise terms to be estimated in (36) are less than those in (18), that is why the model equivalence based recursive extended least squares algorithm has better parameter estimation accuracy.…”

Data filtering-based least squares iterative algorithm for Hammerstein nonlinear systems by using the model decomposition

“…The simulation test validates the effectiveness of the proposed algorithms. The proposed algorithms can be extended to study the parameter estimation problem for dual-rate sampled systems and non-uniformly sampled systems [37][38][39] and Wiener nonlinear systems [40].…”