Abstract
This paper is a continuation of the work in [11] and [2] on the problem of estimating by a linear estimator, N unobservable input vectors, undergoing the same linear transformation, from noise-corrupted observable output vectors. Whereas in the aforementioned papers, only the matrix representing the linear transformation was assumed uncertain, here we are concerned with the case in which the second order statistics of the noise vectors (i.e., their covariance matrices) are also subjected to uncertainty. We seek a robust mean-squared error estimator immuned against both sources of uncertainty. We show that the optimal robust mean-squared error estimator has a special form represented by an elementary block circulant matrix, and moreover when the uncertainty sets are ellipsoidal-like, the problem of finding the optimal estimator matrix can be reduced to solving an explicit semidefinite programming problem, whose size is independent of N.
Similar content being viewed by others
References
Alizadeh, F.: Optimization over the positive-definite cone: interior point methods and combinatorial applications. In: Pardalos, P. (ed.), Advances in optimization and parallel computing. North-Holland, the Netherlands, 1992
Beck, A., Eldar, Y.C., Ben-Tal, A.: Minimax mean-squared error estimation of multichannel signals in the presence of bounded data uncertainties. submitted to SIAM J. Matrix Anal. Appl.
Ben-Haim, Z., Eldar, Y.C.: Maximum set estimators with bounded estimation error. submitted to IEEE Trans. Signal Processing, vol. 53 no. 8, pp. 3172–3182, Aug. 2005
Ben-Haim, Z., Eldar, Y.C.: Minimax estimators dominating the least-squares estimator. submitted to Int. Conf. Acoust., Speech, Signal Processing (ICASSP-2005), pp. 53–56, Mar. 2005
Ben-Tal, A., Nemirovski, A.: Robust convex optimization. Math. Oper. Res. 23 (4), 769–805 (1998)
Ben-Tal, A., Nemirovski, A.: Lectures on modern convex optimization. MPS-SIAM Series on Optimization, 2001
Calafiore, G., El Ghaoui, L.: Robust maximum likelihood estimation in the linear model. Automatica 37, 573–580 (2001)
Chandrasekaran, S., Golub, G.H., Gu, M., Sayed, A.H.: Parameter estimation in the presence of bounded data uncertainties. SIAM J. Matrix Anal. Appl. 19 (1), 235–252 (1998)
Chandrasekaran, S., Gu, M., Sayed, A.H., Schubert, K.E.: The degenerate bounded error-in-variables model. SIAM J. Matrix Anal. Appl. 23 (1), 138–166 (2001)
Eldar, Y.C.: Minimax MSE estimation of deterministic parameters with noise covariance uncertainties. submitted to IEEE Trans. Signal Processing, to appear
Eldar, Y.C., Ben-Tal, A., Nemirovski, A.: Robust mean-squared error estimation in the presence of model uncertainties. IEEE Trans. Signal Processing, vol. 53, pp. 168–181, Jan. 2005
Eldar, Y.C., Ben-Tal, A., Nemirovski, A.: Linear minimax regret estimation of deterministic parameters with bounded data uncertainties. IEEE Trans. Signal Processing 52, 2177–2188 (2004)
Eldar, Y.C., Merhav, N.: Robust linear estimation under a minimax mse–ratio criterion. Technical Report CCIT 449, Technion–Israel Institute of Technology, Haifa, Israel, Oct. 2003. submitted to IEEE Trans. Signal Processing, vol. 52, pp. 2177–2188, Aug. 2004
Eldar, Y.C., Merhav, N.: A competitive minimax approach to robust estimation in linear models. IEEE Trans. Signal Processing 52, 1931–1946 (2004)
Eldar, Y.C., Oppenheim, A.V.: Covariance shaping least-squares estimation. IEEE Trans. Signal Processing 51, 686–697 (2003)
Gauss, K.G.: Theory of motion of heavenly bodies. Dover, New York, NY, 1963
El Ghaoui, L., Lebret, H.: Robust solution to least-squares problems with uncertain data. SIAM J. Matrix Anal. Appl. 18 (4), 1035–1064 (1997)
Golub, G.H., Van Loan, C.F.: An analysis of the total least-squares problem. SIAM J. Numer. Anal. 17 (6), 883–893 (1980)
Golub, G.H., Van Loan, C.F.: Matrix computations. Baltimore MD: Johns Hopkins Univ. Press, 3rd edn., 1996
Hoerl, A.E., Kennard, R.W.: Ridge regression: Biased estimation for nonorthogonal problems. Technometrics 12, 55–67 (1970)
Van Huffel, S., Vandewalle, J.: The total least-squares problem: computational aspects and analysis. Volume 9 of Frontier in Applied Mathematics. Philadelphia, PA: SIAM, 1991
Kailath, T.: Lectures On linear least-squares estimation. Springer, Wein, New York, 1976
Kay, S.M.: Fundamentals of statistical signal processing: estimation theory. Prentice Hall, Inc., Upper Saddle River, NJ, 1993
Mayer, L.S., Willke, T.A.: On biased estimation in linear models. Technometrics, 15, 497–508 (1973)
Nesterov, Y., Nemirovski, A.: Interior-point polynomial algorithms in convex programming. SIAM, Philadelphia, PE, 1994
Sayed, A.H., Nascimento, V., Chandrasekaran, S.: Estimation and control with bounded data uncertainties. Linear Alg. Appl. 284 (1), 259–306 (1998)
Strum, J.F.: Using SeDuMi 1.02, a MATLAB toolbox for optimization over symmetric cones. Optimization methods and software, 11–12, 625–653 (1999)
Therrien, C.W.: Discrete random signals and statistical signal processing. Prentice Hall, Inc., Englewood Cliffs, NJ, 1992
Tikhonov, A.N., Arsenin, V.Y.: Solution of Ill-posed problems. V.H. Winston, Washington, DC, 1977
Vandenberghe, L., Boyd, S.: Semidefinite programming. SIAM Rev. 38 (1), 40–95 (1996)
Author information
Authors and Affiliations
Corresponding author
Additional information
The research was partially supported by BSF grant #2002038
Rights and permissions
About this article
Cite this article
Beck, A., Ben-Tal, A. & Eldar, Y. Robust Mean-Squared Error Estimation of Multiple Signals in Linear Systems Affected by Model and Noise Uncertainties. Math. Program. 107, 155–187 (2006). https://doi.org/10.1007/s10107-005-0683-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10107-005-0683-3
Keywords
- Minimax Mean-Squared Error
- Multiple Observations
- Robust Estimation
- Semidefinite Programming
- Block Circulant Matrices
- Discrete Fourier Transform