Unbiased Recursive Least-Squares Estimation Utilizing Dichotomous Coordinate-Descent Iterations

被引:24
作者
Arablouei, Reza [1 ,2 ]
Dogancay, Kutluyil [1 ,2 ]
Adali, Tulay [3 ]
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] Univ Maryland Baltimore Cty, Dept Comp Sci & Elect Engn, Baltimore, MD 21250 USA
关键词
Adaptive filtering; bias compensation; dichotomous coordinate-descent algorithm; errors-in-variables modeling; recursive least-squares; system identification; PARAMETER-ESTIMATION; BIAS CORRECTION; IDENTIFICATION; VARIABLES; ALGORITHM;
D O I
10.1109/TSP.2014.2316162
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
We propose an unbiased recursive least-squares algorithm for errors-in-variables system identification. The proposed algorithm, called URLS, removes the noise-induced bias when both input and output are contaminated with noise and the input noise is colored and correlated with the output noise. To develop the algorithm, we define an exponentially-weighted least-squares optimization problem that yields an unbiased estimate. Then, we solve the system of linear equations of the associated normal equations utilizing the dichotomous coordinate-descent iterations. The URLS algorithm features significantly reduced computational complexity as well as improved numerical stability compared with a previously proposed bias-compensated recursive least-squares algorithm while having similar estimation performance. We show that the URLS algorithm is asymptotically unbiased and convergent in the mean-square sense. We also calculate its steady-state mean-square deviation. Simulation results corroborate the efficacy of the URLS algorithm and the accuracy of the theoretical findings.
引用
收藏
页码:2973 / 2983
页数:11
相关论文
共 39 条
[1]  
[Anonymous], 1999, SYSTEM IDENTIFICATIO
[2]   Diffusion Bias-Compensated RLS Estimation Over Adaptive Networks [J].
Bertrand, Alexander ;
Moonen, Marc ;
Sayed, Ali H. .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2011, 59 (11) :5212-5224
[3]   AN EFFICIENT RECURSIVE TOTAL LEAST-SQUARES ALGORITHM FOR FIR ADAPTIVE FILTERING [J].
DAVILA, CE .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1994, 42 (02) :268-280
[4]  
DEISTLER M, 1986, J APPL PROBABILITY A, V23, P23
[5]   Bias compensation based recursive least-squares identification algorithm for MISO systems [J].
Ding, Feng ;
Chen, Tongwen ;
Qiu, Li .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2006, 53 (05) :349-353
[6]   Bias compensation-based parameter estimation for output error moving average systems [J].
Ding, Jie ;
Ding, Feng .
INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, 2011, 25 (12) :1100-1111
[7]   A bias-compensated identification approach for noisy FIR models [J].
Diversi, Roberto .
IEEE SIGNAL PROCESSING LETTERS, 2008, 15 :325-328
[9]  
Dogancay K, 2008, PARTIAL-UPDATE ADAPTIVE FILTERS AND ADAPTIVE SIGNAL PROCESSING: DESIGN, ANALYSIS AND IMPLEMENTATION, P1
[10]   Complexity considerations for transform-domain adaptive filters [J].
Dogançay, K .
SIGNAL PROCESSING, 2003, 83 (06) :1177-1192