Improved FOCUSS Method With Conjugate Gradient Iterations
- 31 October 2008
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 57 (1) , 399-404
Abstract
Focal Underdetermined System Solver (FOCUSS) is a powerful tool for sparse representation and underdetermined inverse problems. In this correspondence, we strengthen the FOCUSS method with the following main contributions: 1) we give a more rigorous derivation of the FOCUSS for the sparsity parameter 0 < p < 1 by a nonlinear transform and 2) we develop the CG-FOCUSS by incorporating the conjugate gradient (CG) method to the FOCUSS, which significantly reduces a computational cost with respect to the standard FOCUSS and extends its availability for large scale problems. We justify the CG-FOCUSS based on a probability theory. Furthermore, the high performance of the CG-FOCUSS is demonstrated with experiments.Keywords
This publication has 25 references indexed in Scilit:
- Underdetermined blind source separation based on sparse representationIEEE Transactions on Signal Processing, 2006
- Sparse solutions to linear inverse problems with multiple measurement vectorsIEEE Transactions on Signal Processing, 2005
- Analysis of Sparse Representation and Blind Source SeparationNeural Computation, 2004
- Performance Analysis of Minimum$ell_1$-Norm Solutions for Underdetermined Source SeparationIEEE Transactions on Signal Processing, 2004
- Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ 1 minimizationProceedings of the National Academy of Sciences, 2003
- Subset selection in noise based on diversity measure minimizationIEEE Transactions on Signal Processing, 2003
- Dictionary Learning Algorithms for Sparse RepresentationNeural Computation, 2003
- Underdetermined blind source separation using sparse representationsSignal Processing, 2001
- Atomic Decomposition by Basis PursuitSIAM Journal on Scientific Computing, 1998
- ARPACK Users' Guide1998