Pilanci, M.Arıkan, OrhanPinar, M. C.2016-02-082016-02-082010-10-221053-587Xhttp://hdl.handle.net/11693/22338A 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.EnglishBlind identificationDeconvolutionErrors-in-variablesFrequency estimationLeast squaresRobust least squaresStructured total least squaresBlind identificationBlind identificationsErrors in variablesLeast SquareRobust least squaresStructured total least squaresBlind equalizationCommunication channels (information theory)ConvolutionEstimationMeasurement errorsSignal to noise ratioUncertainty analysisFrequency estimationArticle10.1109/TSP.2010.2041279