Abstract:We study the Compressed Sensing (CS) problem, which is the problem of finding the most sparse vector that satisfies a set of linear measurements up to some numerical tolerance. We introduce an $\ell_2$ regularized formulation of CS which we reformulate as a mixed integer second order cone program. We derive a second order cone relaxation of this problem and show that under mild conditions on the regularization parameter, the resulting relaxation is equivalent to the well studied basis pursuit denoising problem. We present a semidefinite relaxation that strengthens the second order cone relaxation and develop a custom branch-and-bound algorithm that leverages our second order cone relaxation to solve small-scale instances of CS to certifiable optimality. When compared against solutions produced by three state of the art benchmark methods on synthetic data, our numerical results show that our approach produces solutions that are on average $6.22\%$ more sparse. When compared only against the experiment-wise best performing benchmark method on synthetic data, our approach produces solutions that are on average $3.10\%$ more sparse. On real world ECG data, for a given $\ell_2$ reconstruction error our approach produces solutions that are on average $9.95\%$ more sparse than benchmark methods ($3.88\%$ more sparse if only compared against the best performing benchmark), while for a given sparsity level our approach produces solutions that have on average $10.77\%$ lower reconstruction error than benchmark methods ($1.42\%$ lower error if only compared against the best performing benchmark). When used as a component of a multi-label classification algorithm, our approach achieves greater classification accuracy than benchmark compressed sensing methods. This improved accuracy comes at the cost of an increase in computation time by several orders of magnitude.

An Efficient Method for Compressed Sensing

Efficient Compressed Sensing Reconstruction Using Group Sparse Total Variation Regularization

Sparsity-constrained SENSE reconstruction: An efficient implementation using a fast composite splitting algorithm

Compressed Sensing: A Discrete Optimization Approach

Dual-space Compressed Sensing

Compressed Sensing of Complex-Valued Data

Variable Density Compressed Sensing In MRI. Theoretical vs Heuristic Sampling Strategies

Compressive Sensing Based Robust Signal Sampling

Efficient Algorithms for Robust Recovery of Images From Compressed Data

Distributed Convex Optimization Compressed Sensing Method for Sparse Planar Array Synthesis in 3-D Imaging Sonar Systems

Oblique Pursuits for Compressed Sensing

Sparse Signal Reconstruction Via Concave Continuous Piecewise Linear Programming

Fast Compressed Wideband Spectrum Sensing

Compressive Sensing and Its Application: from Sparse to Low-rank Regularized Optimization

Adaptive Compressed Sensing with Diffusion-Based Posterior Sampling

Compressed sensing MRI: a review from signal processing perspective

Optimized Projections for Compressed Sensing Via Direct Mutual Coherence Minimization

Compressive sampling in computed tomography: Method and application

The Implementation and Application Research Development of Compressive Sensing

Image Compressive Sensing Recovery Via Collaborative Sparsity