Performance analysis of the generalised projection identification for time-varying systems

The least mean square methods include two typical parameter estimation algorithms, which are the projection algorithm and the stochastic gradient algorithm, the former is sensitive to noise and the latter is not capable of tracking the time-varying parameters. On the basis of these two typical algor...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IET control theory & applications Jg. 10; H. 18; S. 2506 - 2514
Hauptverfasser: Ding, Feng, Xu, Ling, Zhu, Quanmin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: The Institution of Engineering and Technology 12.12.2016
Schlagworte:
Algorithms
data window length
forgetting factor recursive least squares algorithm
generalised projection algorithm
generalised projection identification algorithm
gradient methods
least mean square methods
Least mean squares algorithm
least mean squares methods
Mathematical models
minimum parameter estimation error
Parameter estimation
parameter estimation algorithms
Parameters
performance analysis
Projection
Research Article
stochastic gradient algorithm
stochastic process theory
stochastic processes
Stochasticity
time‐varying parameters
time‐varying systems
Tracking
time-varying systems
least mean squares methods
minimum parameter estimation error
least mean square methods
stochastic gradient algorithm
time-varying parameters
parameter estimation algorithms
generalised projection identification algorithm
data window length
generalised projection algorithm
parameter estimation
stochastic process theory
stochastic processes
gradient methods
performance analysis
forgetting factor recursive least squares algorithm
ISSN:1751-8644, 1751-8652
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The least mean square methods include two typical parameter estimation algorithms, which are the projection algorithm and the stochastic gradient algorithm, the former is sensitive to noise and the latter is not capable of tracking the time-varying parameters. On the basis of these two typical algorithms, this study presents a generalised projection identification algorithm (or a finite data window stochastic gradient identification algorithm) for time-varying systems and studies its convergence by using the stochastic process theory. The analysis indicates that the generalised projection algorithm can track the time-varying parameters and requires less computational effort compared with the forgetting factor recursive least squares algorithm. The way of choosing the data window length is stated so that the minimum parameter estimation error upper bound can be obtained. The numerical examples are provided.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1751-8644
1751-8652
DOI:10.1049/iet-cta.2016.0202