Compressed Sensing and Affine Rank Minimization Under Restricted Isometry
Penn collection
Degree type
Discipline
Subject
matrix algebra
minimization
affine rank minimization
compressed sensing
constrained l1 minimization
constrained nuclear norm minimization
low-rank matrices
noiseless case
recovery method
restricted isometry conditions
sparse signals
compressed sensing
image reconstruction
noise measurement
signal processing
sparse matrices
vectors
Dantzig selector
compressed sensing
constrained nuclear norm minimization
low-rank matrix recovery
restricted isometry
sparse signal recovery
Computer Sciences
Statistics and Probability
Funder
Grant number
License
Copyright date
Distributor
Related resources
Author
Contributor
Abstract
This paper establishes new restricted isometry conditions for compressed sensing and affine rank minimization. It is shown for compressed sensing that δ<;i>kA<;/i>+θ<;i>k<;/i>,<;i>kA<;/i> <; 1 guarantees the exact recovery of all <;i>k<;/i> sparse signals in the noiseless case through the constrained <;i>l<;/i><;sub>1<;/sub> minimization. Furthermore, the upper bound 1 is sharp in the sense that for any ε > 0, the condition δ<;i>kA<;/i> + θ<;i>k<;/i>,<;i>kA<;/i> <; 1+ε is not sufficient to guarantee such exact recovery using any recovery method. Similarly, for affine rank minimization, if δ<;i>rM<;/i>+θ<;i>r<;/i>,<;i>rM<;/i> <; 1 then all matrices with rank at most <;i>r<;/i> can be reconstructed exactly in the noiseless case via the constrained nuclear norm minimization; and for any ε > 0, δ<;i>rM<;/i> +θ<;i>r<;/i>,<;i>rM<;/i> <; 1+ε does not ensure such exact recovery using any method. Moreover, in the noisy case the conditions δ<;i>kA<;/i>+θ<;i>k<;/i>,<;i>kA<;/i> <; 1 and δ<;i>rM<;/i>+θ<;i>r<;/i>,<;i>rM<;/i> <; 1 are also sufficient for the stable recovery of sparse signals and low-rank matrices respectively. Applications and extensions are also discussed.