Total Least-Squares Algorithm for Wiener Errors-in-Variables System Modeling

Abstract

Abstract: The modeling problem of Wiener errors-in-variables systems was investigated where measurements of the system input and output were corrupted by the additive white Gauss noise. After the provided reformulation of the errors-in-variables system, a two-stage algorithm was developed to estimate the unknown parameters with the first stage employing the total least-squares algorithm, followed by a singular value decomposition in the second stage. The asymptotic maximum likelihood estimation property under the PE condition was strictly proven that with data length tends to infinite, and the proposed total least-squares solution provided an asymptotic maximum likelihood estimate for the nonlinear system parameter vector. The simulation result shows the effectiveness of the proposed algorithm in solving the nonlinear system modeling problem.

Key words: system modeling, errors-in-variables model, Wiener system, total least squares, singular value decomposition

CLC Number:

TP13

Wang Ziyun, Ji Zhicheng. Total Least-Squares Algorithm for Wiener Errors-in-Variables System Modeling[J]. Journal of System Simulation, 2015, 27(8): 1670-1679.

References

[1] Bai E W.A random least-trimmed-squares identification algorithm[J]. Automatica (S0005-1098), 2003, 39(9): 1651-1659.
[2] Soderstrom T.Comparing some classes of bias-compensating least squares methods[J]. Automatica (S0005-1098), 2013, 49(3): 840-845.
[3] Ljung L.System identification-theory for the user[M]. Upper Saddle River, NJ, USA: Prentice Hall, 1999.
[4] Mu B Q, Chen H F.Recursive identification of errors-in-variables Wiener systems[J]. Automatica (S0005-1098), 2013. 49(9): 2744-2753.
[5] Soderstrom T, Irshad Y, Mossberg M, Zheng W X.On the accuracy of a covariance matching method for continuous-time errors-in-variables identification[J]. Automatica (S0005-1098), 2013, 49(10): 2982-2993.
[6] Song Q J.Identification of errors-in-variables systems with nonlinear output observations[J]. Automatica (S0005-1098), 2013, 49(4): 987-992.
[7] Huffel S V.Recent Advances in total least squares techniques and Errors-in-Variables modeling[M]. Philadelphia, PA: SIAM, 1997.
[8] Huffel S V, Lemmerling P.Total least squares and Errors-in-Variables modeling. Analysis, Algorithms and Applications[M]. Dordrecht, The Netherlands: Kluwer, 2002.
[9] Houda S, Samira K.A recursive parametric estimation algorithm of multivariable nonlinear systems described by Hammerstein mathematical models[J]. Applied Mathematical Modelling, 2015, 39(16): 4951-4962.
[10] Han Y H, Raymond A.Hammerstein system identification using nuclear norm minimization. Automatica(S0005-1098), 2012, 48(9): 2189-2193.
[11] Zhao Yanlong, Wang L Y, George G, Zhang J F.Identification of Wiener systems with binary-valued output observations. Automatica(S0005-1098), 2007, 43(10): 1752-1765.
[12] Li J P, Hua C C, Tang Y G, Guan X P.Stochastic gradient with changing forgetting factor-based parameter identification for Wiener systems. Applied Mathematics Letters (S0893-9659), 2014, 33: 40-45.
[13] Yu F, Mao Z Z, Jia M X. Yuan P.Recursive parameter identification of Hammerstein-Wiener systems with measurement noise. Signal Processing (S0165-1684), 2014, 105: 137-147.
[14] Yu F, Mao Z Z, Jia M X.Recursive identification for Hammerstein-Wiener systems with dead-zone input nonlinearity. Journal of Process Control (S0959-1524), 2013, 23(8): 1108-1115.
[15] Mu B Q, Chen H F.Recursive identification of errors-in-variables Wiener-Hammerstein systems. European Journal of Control (S0947-3580), 2014, 20(1): 14-23.
[16] Wills A, Ninness B.Generalised Hammerstein-Wiener system estimation and a benchmark application. Control Engineering Practice (S0967-0661). 2012, 20(11): 1097-1108.
[17] Haryanto A, Hong K.Maximum likelihood identification of Wiener-Hammerstein models. Mechanical Systems and Signal Processing (S0888-3270), 2013, 41(1/2): 54-70.
[18] Wills A, Schön T B, Ljung L, Ninness Brett.Identification of Hammerstein-Wiener models. Automatica(S0005-1098), 2013, 49(1): 70-81.
[19] Schoukens M, Marconato A, Pintelon R, Vandersteen G, Rolain Y.Parametric identification of parallel Wiener- Hammerstein systems. Automatica (S0005-1098), 2015, 51: 111-122.
[20] Schoukens M, Pintelon R, Rolain Y.Identification of Wiener-Hammerstein systems by a nonparametric separation of the best linear approximation. Automatica (S0005-1098), 2014, 50(2): 628-634.
[21] Falck T, Dreesen P, Brabanter D K, Pelckmans K, Moor D B, Suykens J A.K. Least-squares support vector machines for the identification of Wiener-Hammerstein systems. Control Engineering Practice (S0967-0661), 2012, 20(11): 1165-1174.
[22] Bai E W.An optimal two-stage Identification algorithm for Hammerstein-Wiener nonlinear systems[J]. Automatica (S0005-1098), 1998, 34(3): 333-338.
[23] Dolanc G, Strmnik S.Identification of nonlinear systems using a piecewise-linear Hammerstein model. Systems & Control Letters (S0167-6911), 2005, 54(2): 145-158.
[24] Salimifard M, Jafari M, Dehghani M.Identification of nonlinear MIMO block-oriented systems with moving average noises using gradient based and least squares based iterative algorithms. Neurocomputing (S0925-2312), 2012, 94: 22-31.
[25] Chen X, Fang H T.Recursive identification for Hammerstein systems with state-space model. Acta Automatica Sinica (S1874-1029), 2010, 36(10): 1460-1467.
[26] Lee G, Barlow J L.Two projection methods for regularized total least squares approximation. Linear Algebra and its Applications (S0024-3795), 2014, 461: 18-41.
[27] Markovsky I, Huffel V S.Overview of total least-squares methods. Signal Processing (S0165-1684), 2007, 87(10): 2283-2302.
[28] Olsson J, Rydén T.Asymptotic properties of particle filter-based maximum likelihood estimators for state space models. Stochastic Processes and their Applications (S0304-4149), 2008, 118(4): 649-680.
[29] Fujisawa H.Asymptotic properties of conditional maximum likelihood estimator in a certain exponential model. Journal of Multivariate Analysis (S0047-259X), 2003, 86(1): 126-142.
[30] Bibi A, Ghezal A.Consistency of quasi-maximum likelihood estimator for Markov-switching bilinear time series models. Statistics & Probability Letters (S0167-7152), 2015, 100: 192-202.