CHOICE OF STATE-SPACE BASIS IN COMBINED DETERMINISTIC STOCHASTIC SUBSPACE IDENTIFICATION

This paper describes how the state-space basis of models identified wi th subspace identification algorithms can be determined. It is shown t hat this basis is determined by the input spectrum and by user-defined input and output weightings. Through the connections between subspace identification and frequency-weighted balancing, the state-space basi s of the subspace-identified models is shown to coincide with a freque ncy-weighted balanced basis.