TY - JOUR
T1 - Analysis of Regularized LS Reconstruction and Random Matrix Ensembles in Compressed Sensing
AU - Vehkaperä, Mikko
AU - Kabashima, Yoshiyuki
AU - Chatterjee, Saikat
PY - 2016/4/1
Y1 - 2016/4/1
N2 - The performance of regularized least-squares estimation in noisy compressed sensing is analyzed in the limit when the dimensions of the measurement matrix grow large. The sensing matrix is considered to be from a class of random ensembles that encloses as special cases standard Gaussian, row-orthogonal, geometric, and so-called $T$-orthogonal constructions. Source vectors that have non-uniform sparsity are included in the system model. Regularization based on $\ell-{1}$-norm and leading to LASSO estimation, or basis pursuit denoising, is given the main emphasis in the analysis. Extensions to $\ell-{2}$-norm and zero-norm regularization are also briefly discussed. The analysis is carried out using the replica method in conjunction with some novel matrix integration results. Numerical experiments for LASSO are provided to verify the accuracy of the analytical results. The numerical experiments show that for noisy compressed sensing, the standard Gaussian ensemble is a suboptimal choice for the measurement matrix. Orthogonal constructions provide a superior performance in all considered scenarios and are easier to implement in practical applications. It is also discovered that for non-uniform sparsity patterns, the $T$-orthogonal matrices can further improve the mean square error behavior of the reconstruction when the noise level is not too high. However, as the additive noise becomes more prominent in the system, the simple row-orthogonal measurement matrix appears to be the best choice out of the considered ensembles.
AB - The performance of regularized least-squares estimation in noisy compressed sensing is analyzed in the limit when the dimensions of the measurement matrix grow large. The sensing matrix is considered to be from a class of random ensembles that encloses as special cases standard Gaussian, row-orthogonal, geometric, and so-called $T$-orthogonal constructions. Source vectors that have non-uniform sparsity are included in the system model. Regularization based on $\ell-{1}$-norm and leading to LASSO estimation, or basis pursuit denoising, is given the main emphasis in the analysis. Extensions to $\ell-{2}$-norm and zero-norm regularization are also briefly discussed. The analysis is carried out using the replica method in conjunction with some novel matrix integration results. Numerical experiments for LASSO are provided to verify the accuracy of the analytical results. The numerical experiments show that for noisy compressed sensing, the standard Gaussian ensemble is a suboptimal choice for the measurement matrix. Orthogonal constructions provide a superior performance in all considered scenarios and are easier to implement in practical applications. It is also discovered that for non-uniform sparsity patterns, the $T$-orthogonal matrices can further improve the mean square error behavior of the reconstruction when the noise level is not too high. However, as the additive noise becomes more prominent in the system, the simple row-orthogonal measurement matrix appears to be the best choice out of the considered ensembles.
KW - '1 minimization
KW - Compressed sensing
KW - compressed sensing matrices
KW - eigenvalues of random matrices
KW - noisy linear measurements
UR - http://www.scopus.com/inward/record.url?scp=84963759120&partnerID=8YFLogxK
U2 - 10.1109/TIT.2016.2525824
DO - 10.1109/TIT.2016.2525824
M3 - Article
AN - SCOPUS:84963759120
VL - 62
SP - 2100
EP - 2124
JO - IEEE Transactions on Information Theory
JF - IEEE Transactions on Information Theory
SN - 0018-9448
IS - 4
M1 - 7399380
ER -