TY - GEN
T1 - Sparse recovery of multiple measurement vectors in impulsive noise
T2 - 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016
AU - He, Zhen Qing
AU - Shi, Zhi Ping
AU - Huang, Lei
AU - Li, Hongbin
AU - So, H. C.
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/5/18
Y1 - 2016/5/18
N2 - This paper considers the sparse recovery problem of multiple measurement vector (MMV) model corrupted in impulsive noise. To ensure outlier-robust sparse recovery, we formulate an MMV problem that includes the generalized ℓp-norm (1 < p < 2) divergence data-fidelity term added to the ℓ2,0 joint sparsity-promoting regularizer. The joint sparse penalty, however, is non-continuous and hence non-differentiable, which inevitably raises difficulty in optimization when using a gradient-based method. To address this, we build a smooth approximation for the ℓ2,0-based sparse metric via the log-sum based sparse-encouraging surrogate function. Then, we propose a block successive upper-bound minimization algorithm for the smooth MMV problem by solving a series of subproblems based on the block coordinate descent (BCD) method. Furthermore, local convergence of the proposed algorithm to a stationary point of the smooth problem is proved. Experiments demonstrate its efficiency and robust recovery performance for suppressing impulsive noise.
AB - This paper considers the sparse recovery problem of multiple measurement vector (MMV) model corrupted in impulsive noise. To ensure outlier-robust sparse recovery, we formulate an MMV problem that includes the generalized ℓp-norm (1 < p < 2) divergence data-fidelity term added to the ℓ2,0 joint sparsity-promoting regularizer. The joint sparse penalty, however, is non-continuous and hence non-differentiable, which inevitably raises difficulty in optimization when using a gradient-based method. To address this, we build a smooth approximation for the ℓ2,0-based sparse metric via the log-sum based sparse-encouraging surrogate function. Then, we propose a block successive upper-bound minimization algorithm for the smooth MMV problem by solving a series of subproblems based on the block coordinate descent (BCD) method. Furthermore, local convergence of the proposed algorithm to a stationary point of the smooth problem is proved. Experiments demonstrate its efficiency and robust recovery performance for suppressing impulsive noise.
KW - Compressed sensing
KW - impulsive noise
KW - multiple measurement vectors
KW - sparse signal recovery
UR - http://www.scopus.com/inward/record.url?scp=84973300854&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84973300854&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2016.7472537
DO - 10.1109/ICASSP.2016.7472537
M3 - Conference contribution
AN - SCOPUS:84973300854
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 4543
EP - 4547
BT - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings
Y2 - 20 March 2016 through 25 March 2016
ER -