Numerically stable fast sequential calculation for projection approximation subspace tracking

Numerically stable fast sequential calculation for projection approximation subspace tracking

PAST algorithm was developed for subspace tracking from the unconstrained minimization, which resulted in the formula as for RLS adaptive filter with matrix inversion lemma. This paper considers the problem of sequential updating for tracking subspace, which is similar to FTF-type for RLS filtering....

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lee, J., Kyung-Bin Bae
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Adaptive filters
Approximation algorithms
Computer simulation
Direction of arrival estimation
Electronic mail
Filtering
Frequency
Minimization methods
Resonance light scattering
Signal processing algorithms
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PAST algorithm was developed for subspace tracking from the unconstrained minimization, which resulted in the formula as for RLS adaptive filter with matrix inversion lemma. This paper considers the problem of sequential updating for tracking subspace, which is similar to FTF-type for RLS filtering. It is shown that the algorithm has very low complexity, and the performance is investigated through computer simulations. Examples will be given for frequency tracking via ESPRIT.
DOI:10.1109/ISCAS.1999.778888