Yingrou Chen

1. Introduction The Sparse Bayesian learning (SBL) model has been successfully applied to sparse signal recovery (SSR) or image recovery in various fields [1,2,3]. The essence of SSR is to restore the sparse signal x ∈ R N × 1 from the M < N noisy measurement vector y ∈ R M × 1 . In theory, the model is expressed as follows: y = D x + e , (1) where D ∈ R M × N is the known dictionary matrix and e ∈ R M × 1 is the observation noise.