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

被引：88
作者：
Arablouei, Reza [1 ,2 ]
Werner, Stefan [3 ]
Dogancay, Kutluyil [1 ,2 ]
机构：
[1] Univ S Australia, Sch Engn, Mawson Lakes, SA 5095, Australia
[2] Univ S Australia, Inst Telecommun Res, Mawson Lakes, SA 5095, Australia
[3] Aalto Univ, Sch Elect Engn, Dept Signal Proc & Acoust, FI-00076 Aalto, Finland
来源：
IEEE TRANSACTIONS ON SIGNAL PROCESSING | 2014年 / 62卷 / 05期
基金：
芬兰科学院;
关键词：
Adaptive filtering; mean-square deviation; performance analysis; Rayleigh quotient; stability; total least-squares; MINOR COMPONENT ANALYSIS; FIR; CONVERGENCE; ITERATION;
D O I：
10.1109/TSP.2014.2301135
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
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.
引用
收藏
页码：1256 / 1264
页数：9
相关论文
共 43 条
[1]  
Abadir K. M., 2005, MATRIX ALGEBRA, V1
[2]  
Absil PA, 2008, OPTIMIZATION ALGORITHMS ON MATRIX MANIFOLDS, P1
[3]  
[Anonymous], 1997, NUMERICAL LINEAR ALG
[4]  
[Anonymous], 1986, Stability of Adaptive Systems
[5]   Gradient convergence in gradient methods with errors [J].
Bertsekas, DP ;
Tsitsiklis, JN .
SIAM JOURNAL ON OPTIMIZATION, 2000, 10 (03) :627-642
[6]   A TOOL FOR THE ANALYSIS OF QUASI-NEWTON METHODS WITH APPLICATION TO UNCONSTRAINED MINIMIZATION [J].
BYRD, RH ;
NOCEDAL, J .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1989, 26 (03) :727-739
[7]  
Cirrincione G, 2010, ADAPT LEARN SYST SIG, P1, DOI 10.1002/9780470638286
[8]   The MCA EXIN neuron for the minor component analysis [J].
Cirrincione, G ;
Cirrincione, M ;
Hérault, J ;
Van Huffel, S .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (01) :160-187
[9]   AN EFFICIENT RECURSIVE TOTAL LEAST-SQUARES ALGORITHM FOR FIR ADAPTIVE FILTERING [J].
DAVILA, CE .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1994, 42 (02) :268-280
[10]  
Diniz P.S.R., 2013, ADAPTIVE FILTERING A, V4th ed.
12345