Informatica

The aim of the given paper is the development of an approach for parametric identification of Wiener systems with piecewise linear nonlinearities, i.e., when the linear part with unknown parameters is followed by a saturation-like function with unknown slopes. It is shown here that by a simple data reordering and by a following data partition the problem of identification of a nonlinear Wiener system could be reduced to a linear parametric estimation problem. Afterwards, estimates of the unknown parameters of linear regression models are calculated by processing respective particles of input-output data. A technique based on ordinary least squares (LS) is proposed here for the estimation of parameters of linear and nonlinear parts of the Wiener system, including the unknown threshold of piecewise nonlinearity, too. The results of numerical simulation and identification obtained by processing observations of input-output signals of a discrete-time Wiener system with a piecewise nonlinearity by computer are given.

A comparison of two nonlinear input-output models describing the relationship between human emotion (excitement, frustration and engagement/boredom) signals and a virtual 3D face feature (distance-between-eyes) is introduced in this paper. A method of least squares with projection to stability domain for the building of stable models with the least output prediction error is proposed. Validation was performed with seven volunteers, and three types of inputs. The results of the modelling showed relatively high prediction accuracy of excitement, frustration and engagement/boredom signals.

Least-squares method is the most popular method for parameter estimation. It is easy applicable, but it has considerable drawback. Under well-known conditions in the presence of noise, the LS method produces asymptotically biased and inconsistent estimates. One way to overcome this drawback is the implementation of the instrumental variable method. In this paper several modifications of this method for closed-loop system identification are considered and investigated. The covariance matrix of the instrumental variable estimates is discussed. A simulation is carried out in order to illustrate the obtained results.

This paper presents an iterative autoregressive system parameter estimation algorithm in the presence of white observation noise. The algorithm is based on the parameter estimation bias correction approach. We use high order Yule–Walker equations, sequentially estimate the noise variance, and exploit these estimated variances for the bias correction. The improved performance of the proposed algorithm in the presence of white noise is demonstrated via Monte Carlo experiments.

The aim of the given paper is the development of an approach for parametric identification of Hammerstein systems with piecewise linear nonlinearities, i.e., when the saturation-like function with unknown slopes is followed by a linear part with unknown parameters. It is shown here that by a simple input data rearrangement and by a following data partition the problem of identification of a nonlinear Hammerstein system could be reduced to the linear parametric estimation problem. Afterwards, estimates of the unknown parameters of linear regression models are calculated by processing respective particles of input-output data. A technique based on ordinary least squares is proposed here for the estimation of parameters of linear and nonlinear parts of the Hammerstein system, including the unknown threshold of the piecewise nonlinearity, too. The results of numerical simulation and identification obtained by processing observations of input-output signals of a discrete-time Hammerstein system with a piecewise nonlinearity with positive slopes by computer are given.