ISSN:
1573-2878
Schlagwort(e):
Set membership estimation
;
linear models
;
projection algorithms
;
conditional estimation
;
worst-case error
Quelle:
Springer Online Journal Archives 1860-2000
Thema:
Mathematik
Notizen:
Abstract This paper studies the role of projection algorithms in conditional set membership estimation. These algorithms are known to be suboptimal in terms of the worst-case estimation error. A tight upper bound on the error of central projection estimators and interpolatory projection estimators is computed as a function of the conditional radius of information. Since the radius of information represents the minimum achievable error, the derived bound provides a measure of the reliability level of the suboptimal algorithms. The results are derived in a general deterministic setting, which allows the consideration of linearly parametrized approximations of a compact set of feasible problem elements.
Materialart:
Digitale Medien
URL:
http://dx.doi.org/10.1023/A:1021710825323
Permalink
