ULV and generalized ULV subspace tracking adaptive algorithms

Srinath Hosur, Ahmed H. Tewfik, Daniel Boley

Research output: Contribution to journalArticlepeer-review

Abstract

Traditional adaptive filters assume that the effective rark of the input signal is the same as the input covariance matrix or the filter length N. Therefore, if the input signal lives in a subspace of dimension less than N, these filters fail to perform satisfactorily. In this paper, we present two new algorithms for adapting only in the dominant signal subspace. The first of these is a low-rank recursive-least-squares (RLS) algorithm that uses a ULY decomposition to track and adapt in the signal subspace. The second adaptive algorithm is a subspace tracking leastmean-squares (LMS) algorithm that uses a generalized ULV (GULV) decomposition, which has been developed in this paper, to track and adapt in subspaces corresponding to several well-conditioned singular cue clusters. The algorithm also has an improved convergence speed compared with that of the LMS algorithm. Bounds on the quality of subspaces isolated using the GULV decomposition are derived, and the performance of the adaptive algorithms are analyzed.

Original languageEnglish (US)
Number of pages1
JournalIEEE Transactions on Signal Processing
Volume45
Issue number1
StatePublished - Dec 1 1997

Fingerprint Dive into the research topics of 'ULV and generalized ULV subspace tracking adaptive algorithms'. Together they form a unique fingerprint.

Cite this