Frequency-domain subspace system identification with fractional differentiation models

In this paper, an algorithm based on subspace method from the MOESP (MIMO Ouput-Erreur State sPace) family is presented, for state-space identification of continuous-time fractional commensurate models, using samples of frequency data. As compared to the rational state-space representation, the commensurate differentiation order must be estimated besides the state-space matrices, estimated with conventional subspace-based techniques using QR and singular value decomposition. This is the first method developed for multi-input multi-output system identification of fractional models in the frequency domain.