Title :
A Robust Algorithm for Joint Sparse Recovery in Presence of Impulsive Noise
Author :
Jiadong Shang ; Zulin Wang ; Qin Huang
Author_Institution :
Sch. of Electr. & Inf. Eng., Beihang Univ., Beijing, China
Abstract :
This letter presents a robust solution for joint sparse recovery (JSR) under impulsive noise. The unknown measurement noise is endowed with the Student-t distribution, then a novel Bayesian probabilistic model is proposed to describe the JSR problem. To effectively recover the joint row sparse signal, variational Bayes (VB) method is introduced for Bayesian theory based JSR algorithms such that it overcomes the intractable integrations inherent. Simulation results verify that the proposed algorithm significantly outperforms the existing algorithms under impulsive noise.
Keywords :
Bayes methods; compressed sensing; Bayesian theory; JSR; Student-t distribution; VB method; impulsive noise; joint sparse recovery; noise measurement; novel Bayesian probabilistic model; robust algorithm; variational Bayes method; Approximation algorithms; Bayes methods; Joints; Noise; Noise measurement; Robustness; Signal processing algorithms; Bayesian inference; impulsive noise; joint sparse recovery; student-t distribution; variational Bayes method;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2014.2387435
