Inertial projection neural network for nonconvex sparse signal recovery with prior information

1 Introduction In this letter,we consider the problem of recovering an unknown sparse signal from its few measurements when part of the support prior of the desired signal is available.This problem naturally raises from the compressed sensing(CS),which was proposed by Candès et al.[1]around 2006.Since then,CS has rapidly developed,see,e.g.,[2-5].