A neural network for l1 - l2 minimization based on scaled gradient projection: Application to compressed sensing

Since compressed sensing was introduced in 2006, l(1) - l(2) minimization admits a large number of applications in signal processing, statistical inference, magnetic resonance imaging (Mm), computed tomography (CT), etc. In this paper, we present a neural network for l(1) - l(2) minimization based on scaled gradient projection. We prove that it is stable in the sense of Lyapunov and converges to an optimal solution of the l(1) - l(2) minimization. We show that the proposed neural network is feasible and efficient for compressed sensing via simulation examples. (C) 2015 Elsevier B.V. All rights reserved.