Model-set design for multiple-model method. Part I

The most important problem in the application of the multiple-model approach to estimation is the design of the model set. This paper deals with this challenging topic in a general setting. Modeling of models as well as true mode as random variables is proposed. Several general methods for design of model sets, along with the initial model probabilities, are presented. They include distribution approximation, minimizing mismatch between mode and models, and moment matching. Examples that demonstrate how the general results presented here can be applied are presented in Part II.