Statistical efficiency of structured CPD estimation applied to Wiener-Hammerstein modeling

The computation of a structured canonical polyadic decomposition (CPD) is useful to address several important modeling problems in real-world applications. In this paper, we consider the identification of a nonlinear system by means of a Wiener-Hammerstein model, assuming a high-order Volterra kernel of that system has been previously estimated. Such a kernel, viewed as a tensor, admits a CPD with banded circulant factors which comprise the model parameters. To estimate them, we formulate specialized estimators based on recently proposed algorithms for the computation of structured CPDs. Then, considering the presence of additive white Gaussian noise, we derive a closed-form expression for the Cramér-Rao bound (CRB) associated with this estimation problem. Finally, we assess the statistical performance of the proposed estimators via Monte Carlo simulations, by comparing their mean-square error with the CRB.

[1]  Gérard Favier,et al.  An Algebraic Solution for the Candecomp/PARAFAC Decomposition with Circulant Factors , 2014, SIAM J. Matrix Anal. Appl..

[2]  João Cesar M. Mota,et al.  Blind channel identification algorithms based on the Parafac decomposition of cumulant tensors: The single and multiuser cases , 2008, Signal Process..

[3]  J. Nocedal,et al.  A Limited Memory Algorithm for Bound Constrained Optimization , 1995, SIAM J. Sci. Comput..

[4]  Souleymen Sahnoun,et al.  Tensor polyadic decomposition for antenna array processing , 2014 .

[5]  Louis L. Scharf,et al.  Signal processing applications of oblique projection operators , 1994, IEEE Trans. Signal Process..

[6]  Pierre Comon,et al.  Blind Multilinear Identification , 2012, IEEE Transactions on Information Theory.

[7]  Jorge Nocedal,et al.  A Limited Memory Algorithm for Bound Constrained Optimization , 1995, SIAM J. Sci. Comput..

[8]  Lieven De Lathauwer,et al.  Blind Signal Separation via Tensor Decomposition With Vandermonde Factor: Canonical Polyadic Decomposition , 2013, IEEE Transactions on Signal Processing.

[9]  Pierre Comon,et al.  Performance estimation for tensor CP decomposition with structured factors , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[10]  Robert Haber.  Nonlinear System Identification : Input-output Modeling Approach , 1999 .

[11]  Pierre Comon,et al.  Tensors : A brief introduction , 2014, IEEE Signal Processing Magazine.

[12]  Ieee Staff.  2017 25th European Signal Processing Conference (EUSIPCO) , 2017 .

[13]  A.Y. Kibangou,et al.  Wiener-Hammerstein systems modeling using diagonal Volterra kernels coefficients , 2006, IEEE Signal Processing Letters.

[14]  G KoldaTamara,et al.  Tensor Decompositions and Applications , 2009 .

[15]  Nikos D. Sidiropoulos,et al.  Cramer-Rao lower bounds for low-rank decomposition of multidimensional arrays , 2001, IEEE Trans. Signal Process..

[16]  Gérard Favier,et al.  Non-iterative solution for PARAFAC with a Toeplitz matrix factor , 2009, 2009 17th European Signal Processing Conference.

[17]  S. Kay.  Fundamentals of statistical signal processing: estimation theory , 1993 .

[18]  A. Kibangou,et al.  Identification of fifth-order Volterra systems using i.i.d. inputs , 2010 .

[19]  Gérard Favier,et al.  Tensor-based methods for system identification. Part 2: Three examples of tensor-based system identification methods , 2009 .

[20]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[21]  Pierre Comon,et al.  Tensor decompositions with banded matrix factors , 2013 .