Method of performing structure-based bayesian sparse signal reconstruction

摘要

The method of performing structure-based Bayesian sparse signal reconstruction is a Bayesian approach to sparse signal recovery that has relatively low complexity and makes a collective use of: a priori statistical properties of the signal and noise; sparsity information; and the rich structure of the sensing matrix Ψ. The method is used with both Gaussian and non-Gaussian (or unknown) priors, and performance measures of the ultimate estimate outputs are easily calculated.