Nonparametric identification of nonlinear biomedical systems, part I: Theory

The construction of mathematical models of physiological systems based on experimental data is a classic example of a nonlinear system identification problem. This article reviews theoretical and practical issues relevant to nonlinear system identification in the context of biomedical engineering. Various model structures, commonly used to describe nonlinear physiological systems, are described using a common theoretical framework and notation, to elucidate relationships among them. We then review methods for the identification of the different models using the common theoretical perspective, previously developed, and compare and contrast the strengths and weaknesses of the various analysis methods in use.