Choice of state-space basis in combined deterministic-stochastic subspace identification (Q1911280)

The paper gives a general presentation of the identification problem of linear and stochastic systems. It is shown that various algorithms differ in a weighting only. The main theorem introduces specific choices for the weighting matrices, such that the system is identified in a predefined state space basis. A space-state interpretation of frequency-weighted model reduction (Enns, 1984) is given, too.