Abstract
In this paper the problem of identifying noisy input-output systems is studied. The paper revisits the Koopmans-Levin (KL) method by investigating its sensitivity with respect to the ratio between the variances of the input noise and the output noise. In particular, the sensitivity analysis suggests that the assumption of the given noise variance ratio can be relaxed to that of a rough range of this ratio. Thus the practical applicability of the KL method is greatly extended. Computer simulations confirm the theoretical predictions.
| Original language | English |
|---|---|
| Pages (from-to) | 636-637 |
| Number of pages | 2 |
| Journal | Proceedings of the IEEE Conference on Decision and Control |
| Volume | 1 |
| Publication status | Published - 2002 |
| Event | 41st IEEE Conference on Decision and Control - Las Vegas, NV, United States Duration: 10 Dec 2002 → 13 Dec 2002 |