Visible to the public Analysis of the Gradient-Descent Total Least-Squares Adaptive Filtering Algorithm

Submitted by BrandonB on Wed, 05/06/2015 - 9:27am
TitleAnalysis of the Gradient-Descent Total Least-Squares Adaptive Filtering Algorithm
Publication TypeJournal Article
Year of Publication2014
AuthorsArablouei, R., Werner, S., Dogancay, K.
JournalSignal Processing, IEEE Transactions on
Volume62
Pagination1256-1264
Date PublishedMarch
ISSN1053-587X
Keywordsadaptive filtering, adaptive filters, Algorithm design and analysis, energy-conservation, gradient-descent total least-squares algorithm, Least squares approximations, mean-square deviation, Performance analysis, Rayleigh quotient, Signal processing algorithms, stability, Stability criteria, Steady-state, steady-state analysis, steady-state mean-square deviation, Stochastic processes, stochastic-gradient adaptive filtering algorithm, total least-squares, Vectors
Abstract

The gradient-descent total least-squares (GD-TLS) algorithm is a stochastic-gradient adaptive filtering algorithm that compensates for error in both input and output data. We study the local convergence of the GD-TLS algoritlun and find bounds for its step-size that ensure its stability. We also analyze the steady-state performance of the GD-TLS algorithm and calculate its steady-state mean-square deviation. Our steady-state analysis is inspired by the energy-conservation-based approach to the performance analysis of adaptive filters. The results predicted by the analysis show good agreement with the simulation experiments.
URLhttp://ieeexplore.ieee.org/document/6716043/
DOI10.1109/TSP.2014.2301135
Citation Key6716043
Groups: