A smoothing neural network for minimization l-l in sparse signal reconstruction with measurement noises.

Journal: Neural networks : the official journal of the International Neural Network Society
Published Date:

Abstract

This paper investigates a smoothing neural network (SNN) to solve a robust sparse signal reconstruction in compressed sensing (CS), where the objective function is nonsmooth l-norm and the feasible set satisfies an inequality of l-norm 2≥p≥1 which is used for measuring residual errors. With a smoothing approximate technique, the non-smooth and non-Lipschitz continuous issues of the l-norm and the gradient of l-norm can be solved efficiently. We propose a SNN which is modeled by a differential equation and give its circuit implementation. In this case, we prove the proposed SNN converges to the optimal of considered problem. Simulation results are discussed to demonstrate the efficiency of the proposed algorithm.

Authors

  • You Zhao
    National & Local Joint Engineering Laboratory of Intelligent Transmission and Control Technology (Chongqing), College of Electronic and Information Engineering, Southwest University, Chongqing, 400715, China; Key laboratory of Machine Perception and Children's Intelligence Development, Chongqing University of Education, Chongqing, 400067, China. Electronic address: Zhaoyou1991sdtz@163.com.
  • Xing He
    University of Florida, Gainesville, Florida, USA.
  • Tingwen Huang
  • Junjian Huang
    Department of Computer Science, Chongqing University of Education, Chongqing 400067, China. Electronic address: hmomu@sina.com.
  • Peng Li
    WuXi AppTec Co, Shanghai, China.