Study of the minimum experiment length to identify linear dynamic systems: a variance based approach.

摘要

In this paper the effect of short data lengths in system identification is studied. It addresses the question of the minimum required data length that is needed in order to apply the asymptotic results on the uncertainty analysis. this paper is focused on the IIR-case by analyzing initially a first order system. The conclusions are extended to higher order systems by normalizing all results on the time constant of this system, and by adding a model complexity factor.