Comments On`state -space Model Identiication with Data Correlation'

In a recent paper, Hou and Hsu (1991) derive a state-space identii-cation method which apparently improves upon the results in (Moonen et al. 1989). Here we point out that neither one of these methods really applies to the examples given in the paper, and we give an outline of a method which should be used instead. In their paper, Hou and Hsu (1991) consider general state space models of the form x k+1 = A x k + B u k + w k y k = C x k + D u k + v k where y k 2 < l is the observed output and u k 2 < m is the observed determin-istic input. The process noise w k and measurement noise v k are unknown. The aim is to identify the system matrices A; B; C; D {up to a similarity transformation{ by means of recorded I/O-sequences fu 1 u 2 : : : u N g and fy 1 y 2 : : : y N g.