Convergence of hierarchical stochastic gradient identification for transfer function matrix model (Q2783698)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Convergence of hierarchical stochastic gradient identification for transfer function matrix model |
scientific article; zbMATH DE number 1730669
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Convergence of hierarchical stochastic gradient identification for transfer function matrix model |
scientific article; zbMATH DE number 1730669 |
Statements
15 December 2002
0 references
hierarchical stochastic gradient procedure
0 references
identification
0 references
linear discrete-time MIMO stochastic systems
0 references
martingale hyperconvergence theorem
0 references
Convergence of hierarchical stochastic gradient identification for transfer function matrix model (English)
0 references
The paper is concerned with the parametric identification of a linear discrete-time MIMO stochastic system described by a transfer function matrix model. The system is decomposed into two fictitious partial subsystems with, respectively, vector and matrix parameters. To estimate these parameters, the recursive hierarchical stochastic gradient identification procedure is proposed and examined. It is shown that, for persistently exciting input, the algorithm converges to the true values of system parameters with probability one. The convergence analysis is based on the martingale hyperconvergence theorem.
0 references
