Abstract
In this paper, we consider the identification problem for a dual-rate system in which the input sampling period may differ from that of the output. Based on the lifting operators, a lifted system which is equivalent to the original dual-rate system can be derived so that a lifted state-space model can be obtained which maps the relations between the dual-rate input-output data. Then the Numerical Subspace State-Space IDentification (N4SID) algorithm is modified and used to identify the lifted state-space model for the first time in the literature, taking the causality constraints of the lifted system into account. Finally, numerical studies are included to show the excellent numerical performance of the proposed algorithm.
