Article

Structured least squares with bounded data uncertainties

ABSTRACT In many signal processing applications the core problem reduces to a linear system of equations. Coefficient matrix uncertainties create a significant challenge in obtaining reliable solutions. In this paper, we present a novel formulation for solving a system of noise contaminated linear equations while preserving the structure of the coefficient matrix. The proposed method has advantages over the known Structured Total Least Squares (STLS) techniques in utilizing additional information about the uncertainties and robustness in ill-posed problems. Numerical comparisons are given to illustrate these advantages in two applications: signal restoration problem with an uncertain model and frequency estimation of multiple sinusoids embedded in white noise.

Download full-text

Full-text

Available from: Orhan Arikan, Jul 03, 2015
0 Followers
 · 
82 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: In many inverse problems of signal processing the problem reduces to a linear system of equations. Accurate and robust estimation of the solution with errors in both measurement vector and coefficient matrix is a challenging task. In this paper a novel formulation is proposed which takes into account the structure (e.g. Toeplitz, Hankel) and uncertainties of the system. A numerical algorithm is provided to obtain the solution. The proposed technique and other methods are compared in a channel equalization example which is a fundamental necessity in communication.
    Signal Processing and Communications Applications Conference, 2009. SIU 2009. IEEE 17th; 05/2009
  • [Show abstract] [Hide abstract]
    ABSTRACT: In many signal processing problems such as channel estimation and equalization, the problem reduces to a linear system of equations. In this proceeding we formulate and investigate linear equations systems with sparse perturbations on the coefficient matrix. In a large class of matrices, it is possible to recover the unknowns exactly even if all the data, including the coefficient matrix and observation vector is corrupted. For this aim, we propose an optimization problem and derive its convex relaxation. The numerical results agree with the previous theoretical findings of the authors. The technique is applied to adaptive multipath estimation in cognitive radios and a significant performance improvement is obtained. The fact that rapidly varying channels are sparse in delay and doppler domain enables our technique to maintain reliable communication even far from the channel training intervals.
  • [Show abstract] [Hide abstract]
    ABSTRACT: A novel approach is proposed to provide robust and accurate estimates for linear regression problems when both the measurement vector and the coefficient matrix are structured and subject to errors or uncertainty. A new analytic formulation is developed in terms of the gradient flow of the residual norm to analyze and provide estimates to the regression. The presented analysis enables us to establish theoretical performance guarantees to compare with existing methods and also offers a criterion to choose the regularization parameter autonomously. Theoretical results and simulations in applications such as blind identification, multiple frequency estimation and deconvolution show that the proposed technique outperforms alternative methods in mean-squared error for a significant range of signal-to-noise ratio values.
    IEEE Transactions on Signal Processing 06/2010; DOI:10.1109/TSP.2010.2041279 · 3.20 Impact Factor