English

A Robust Deep Unfolded Network for Sparse Signal Recovery from Noisy Binary Measurements

Machine Learning 2020-10-16 v1 Signal Processing

Abstract

We propose a novel deep neural network, coined DeepFPC-2\ell_2, for solving the 1-bit compressed sensing problem. The network is designed by unfolding the iterations of the fixed-point continuation (FPC) algorithm with one-sided 2\ell_2-norm (FPC-2\ell_2). The DeepFPC-2\ell_2 method shows higher signal reconstruction accuracy and convergence speed than the traditional FPC-2\ell_2 algorithm. Furthermore, we compare its robustness to noise with the previously proposed DeepFPC network---which stemmed from unfolding the FPC-1\ell_1 algorithm---for different signal to noise ratio (SNR) and sign-flipped ratio (flip ratio) scenarios. We show that the proposed network has better noise immunity than the previous DeepFPC method. This result indicates that the robustness of a deep-unfolded neural network is related with that of the algorithm it stems from.

Keywords

Cite

@article{arxiv.2010.07564,
  title  = {A Robust Deep Unfolded Network for Sparse Signal Recovery from Noisy Binary Measurements},
  author = {Y. Yang and P. Xiao and B. Liao and N. Deligiannis},
  journal= {arXiv preprint arXiv:2010.07564},
  year   = {2020}
}

Comments

5 pages, 5 figures, conference

R2 v1 2026-06-23T19:22:02.292Z