In this paper we derive weighted and reweighted AMP algorithms for signal reconstruction from compressed sensing measurements. Weighted AMP incorporates prior support information into the AMP algorithm and iteratively solves the weighted I1 minimization which is much faster than the usual linear programming algorithms used to solve this problem. We also introduce a reweighting scheme for regular and weighted AMP algorithms which enhances the recovery performance of both regular and weighted AMP while still maintaining the low complexity nature of AMP algorithms.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Navid Ghadermarzy and Ozgur Yilmaz
Weighted and reweighted approximate message passing
", Proc. SPIE 8858, Wavelets and Sparsity XV, 88580C (September 26, 2013); doi:10.1117/12.2027069; http://dx.doi.org/10.1117/12.2027069