TY - JOUR
T1 - Robust one-bit bayesian compressed sensing with sign-flip errors
AU - Li, Fuwei
AU - Fang, Jun
AU - Li, Hongbin
AU - Huang, Lei
N1 - Publisher Copyright:
© 1994-2012 IEEE.
PY - 2015/7/1
Y1 - 2015/7/1
N2 - We consider the problem of sparse signal recovery from one-bit measurements. Due to the noise present in the acquisition and transmission process, some quantized bits may be flipped to their opposite states. These bit-flip errors, also referred to as the sign-flip errors, may result in severe performance degradation. To address this issue, we introduce a robust Bayesian compressed sensing framework to account for sign flip errors. Specifically, sign-flip errors are considered as a result of a sparse noise-corrupted model in which original (unquantized) observations are corrupted by sparse (impulse) noise. A Gaussian-inverse Gamma hierarchical prior is assigned to the noise vector to promote sparsity. Based on the modified hierarchical model, we develop a variational expectation-maximization (EM) algorithm to identify the sign-flip errors and recover the sparse signal simultaneously. Numerical results are provided to illustrate the effectiveness and superiority of the proposed method.
AB - We consider the problem of sparse signal recovery from one-bit measurements. Due to the noise present in the acquisition and transmission process, some quantized bits may be flipped to their opposite states. These bit-flip errors, also referred to as the sign-flip errors, may result in severe performance degradation. To address this issue, we introduce a robust Bayesian compressed sensing framework to account for sign flip errors. Specifically, sign-flip errors are considered as a result of a sparse noise-corrupted model in which original (unquantized) observations are corrupted by sparse (impulse) noise. A Gaussian-inverse Gamma hierarchical prior is assigned to the noise vector to promote sparsity. Based on the modified hierarchical model, we develop a variational expectation-maximization (EM) algorithm to identify the sign-flip errors and recover the sparse signal simultaneously. Numerical results are provided to illustrate the effectiveness and superiority of the proposed method.
KW - One-bit Bayesian compressed sensing
KW - sign-flip errors
KW - variational expectation-maximization
UR - http://www.scopus.com/inward/record.url?scp=84916608742&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84916608742&partnerID=8YFLogxK
U2 - 10.1109/LSP.2014.2373380
DO - 10.1109/LSP.2014.2373380
M3 - Article
AN - SCOPUS:84916608742
SN - 1070-9908
VL - 22
SP - 857
EP - 861
JO - IEEE Signal Processing Letters
JF - IEEE Signal Processing Letters
IS - 7
M1 - 6963346
ER -