On discretization of continuous-time state-space models: a stable-normal approach

The problem of developing discrete-time models for continuous-time systems from some closed-form discretizing transformations is discussed. Stable forms of normal interpolators and integrating operators, which are of key issue in an associated normal class of transformations, are given, and their approximating properties are shown in the frequency domain. Generic forms of state-space models obtained by a simple piece-wise-normal rule (time invariance constrained to one sampling period) and by direct and indirect stable-normal approximation methods (originally resulting in a discrete-time extended state-space model) are developed, and numerical examples are given.<>