Numerical conditioning and asymptotic variance of subspace estimates

被引：16

作者：

Chiuso, A ^{[1
]}

Picci, G ^{[1
]}

机构：

[1] Univ Padua, Dept Informat Engn, I-35131 Padua, Italy

来源：

AUTOMATICA | 2004年 / 40卷 / 04期

关键词：

asymptotic variance; subspace identification; exogenous inputs; numerical conditioning; collinearity; oblique projections; state-space identification;

D O I：

10.1016/j.automatica.2003.11.008

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

New formulas for the asymptotic variance of the parameter estimates in subspace identification, show that the accuracy of the parameter estimates depends on certain indices of 'near collinearity' of the state and future input subspaces of the system to be identified. This complements the numerical conditioning analysis of subspace methods presented in the companion paper (On the ill-conditioning of subspace identification with inputs, Automatica, doi:10.1016/j.automatica.2003.11.009). (C) 2003 Elsevier Ltd. All rights reserved.

引用

页码：677 / 683

页数：7

共 10 条

[1] Analysis of the asymptotic properties of the MOESP type of subspace algorithms [J].