Convergence of stochastic gradient estimation algorithm for multivariable ARX-like systems

A New Identification Approach of MIMO Hammerstein Model with Separate Nonlinearities

“…They can be divided into stochastic methods [6,7], iterative methods [8], over-parameterization methods [9], separable least squares methods [10,11], blind identification methods [12] and frequency domain methods [13]. It is possible to transform the SISO Hammerstein model to MISO and MIMO model which is linear in the parameters [14]. Serval approches have been proposed to identify MIMO Hammerstein model.…”

Dynamic modeling of breast tissue with application of model reference adaptive system identification technique based on clinical robot-assisted palpation

“…A recursive least squares estimation was proposed for output error moving average systems using data filtering [20]. In the area of parameter estimation [21][22][23][24], Liu et al proposed a gradient approach for multiple-input and singleoutput systems using identification theory and the auxiliary model identification idea [25] and analyzed the convergence of the gradient algorithm for multivariable Autoregressive with exogenous inputs (ARX )-like systems [26]. Xiao et al presented a residual based interactive least squares algorithm for controlled autoregressive moving average (ARMA) systems [27].…”

Identification of MIMO Hammerstein models using Singular Value Decomposition approach

“…Although there are some methods that can handle MIMO processes [22], many of them assume the structure of the nonlinearity to be separate [28], i.e., the i th output of the nonlinear function is only affected by the i th input. neural networks or fuzzy logic [1,2,14,15], and polynomial with cross-terms have often been used to deal with more general nonlinearities.…”

“…Recently, several approaches have been proposed to identify MIMO Hammerstein models. For examples, Liu et al in [22] extended Sung's [29] decoupling method to MIMO systems, which however requires nonlinear optimization. Kwong et al in [18] extended Ramos's [28] method to MIMO models.…”