Detecting Instability Potentials in Regularization for Fast Affine Projection Algorithms
In fast affine projection (FAP) adaptation algorithms, it is needed to explicitly or implicitly perform a matrix inversion, during which a small positive regularization factor plays an important role in keeping the algorithm stable and optimized. While existing schemes choose the regularization fact...
Saved in:
Published in | 2007 Conference Record of the Forty-First Asilomar Conference on Signals, Systems and Computers pp. 1623 - 1627 |
---|---|
Main Author | |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
01.11.2007
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | In fast affine projection (FAP) adaptation algorithms, it is needed to explicitly or implicitly perform a matrix inversion, during which a small positive regularization factor plays an important role in keeping the algorithm stable and optimized. While existing schemes choose the regularization factor based on certain system criteria not related to the inversion, this paper proposes a simple scheme that dynamically diagnoses the inversion process itself for potentials of instability. This work paves the way for further studies on "minimal regularization and step-size control" technique. A FAP adopting this technique can be compared with FAPs with existing regularization schemes for convergence and steady state performance. |
---|---|
ISBN: | 9781424421091 1424421098 |
ISSN: | 1058-6393 2576-2303 |
DOI: | 10.1109/ACSSC.2007.4487506 |