Sparsity-Cognizant Total Least-Squares for Perturbed Compressive Sampling
From MaRDI portal
Publication:4572912
Abstract: Solving linear regression problems based on the total least-squares (TLS) criterion has well-documented merits in various applications, where perturbations appear both in the data vector as well as in the regression matrix. However, existing TLS approaches do not account for sparsity possibly present in the unknown vector of regression coefficients. On the other hand, sparsity is the key attribute exploited by modern compressive sampling and variable selection approaches to linear regression, which include noise in the data, but do not account for perturbations in the regression matrix. The present paper fills this gap by formulating and solving TLS optimization problems under sparsity constraints. Near-optimum and reduced-complexity suboptimum sparse (S-) TLS algorithms are developed to address the perturbed compressive sampling (and the related dictionary learning) challenge, when there is a mismatch between the true and adopted bases over which the unknown vector is sparse. The novel S-TLS schemes also allow for perturbations in the regression matrix of the least-absolute selection and shrinkage selection operator (Lasso), and endow TLS approaches with ability to cope with sparse, under-determined "errors-in-variables" models. Interesting generalizations can further exploit prior knowledge on the perturbations to obtain novel weighted and structured S-TLS solvers. Analysis and simulations demonstrate the practical impact of S-TLS in calibrating the mismatch effects of contemporary grid-based approaches to cognitive radio sensing, and robust direction-of-arrival estimation using antenna arrays.
Cited in
(24)- A study on off-grid issue in DOA and frequency estimations
- Robust sparse signal reconstructions against basis mismatch and their applications
- Sparse total least squares: analysis and greedy algorithms
- A convex optimization framework for the identification of homogeneous reaction systems
- Sparse linear regression from perturbed data
- Sign patterns of inverse doubly-nonnegative matrices
- Off-grid DOA estimation via real-valued sparse Bayesian method in compressed sensing
- Signal recovery and system calibration from multiple compressive Poisson measurements
- Modified-CS: Modifying Compressive Sensing for Problems With Partially Known Support
- Fully Decomposable Compressive Sampling With Joint Optimization for Multidimensional Sparse Representation
- Sparse Online Variational Bayesian Regression
- Representation and reconstruction of covariance operators in linear inverse problems
- Solving inverse problems using data-driven models
- Toeplitz Compressed Sensing Matrices With Applications to Sparse Channel Estimation
- Plug in estimation in high dimensional linear inverse problems a rigorous analysis
- Regularized TLS-EM for estimating missing data
- Sparse Bayesian perspective for radar coincidence imaging with model errors
- A Levenberg-Marquardt method for nonsmooth regularized least squares
- CR-Lasso: robust cellwise regularized sparse regression
- A sparse reconstruction algorithm based on constrained inhomogeneous grid optimization
- Maximum likelihood inference for high-dimensional problems with multiaffine variable relations
- A Bayesian approach to total least-squares in perturbed compressive sensing
- Covariate Selection in High-Dimensional Generalized Linear Models With Measurement Error
- Measurement matrix design for CS-MIMO radar using multi-objective optimization
This page was built for publication: Sparsity-Cognizant Total Least-Squares for Perturbed Compressive Sampling
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4572912)
