CG-M-FOCUSS and Its Application to Distributed Compressed Sensing

Abstract. M-FOCUSS is one of themost successful and efficientmethods for sparse representation. To reduce the computational cost ofM-FOCUSS and to extend its availability for large scale problems, M-FOCUSS is extended to CG-M-FOCUSS by incorporating conjugate gradient (CG) iterations in this paper. Furthermore, the CG-M-FOCUSS is applied to distributed compressed sensing. We illustrate the performance of CG-M-FOCUSS by an MRI image reconstruction example, in which CG-M-FOCUSS can not only reconstruct theMRI image with high precision, but also considerably reduce the computational time