Bayesian signal reconstruction for 1-bit compressed sensing

Abstract

The 1-bit compressed sensing framework enables the recovery of a sparse vector x from the sign information of each entry of its linear transformation. Discarding the amplitude information can significantly reduce the amount of data, which is highly beneficial in practical applications. In this paper, we present a Bayesian approach to signal reconstruction for 1-bit compressed sensing, and analyze its typical performance using statistical mechanics. Utilizing the replica method, we show that the Bayesian approach enables better reconstruction than the L1-norm minimization approach, asymptotically saturating the performance obtained when the non-zero entries positions of the signal are known. We also test a message passing algorithm for signal reconstruction on the basis of belief propagation. The results of numerical experiments are consistent with those of the theoretical analysis.

24pages,9figures

Cited by in corpus (9)

• Statistical physics of inference: Thresholds and algorithms
• Bayes-Optimal Joint Channel-and-Data Estimation for Massive MIMO with Low-Precision ADCs
• Optimal Errors and Phase Transitions in High-Dimensional Generalized Linear Models
• Nonconvex penalties with analytical solutions for one-bit compressive sensing
• Generalized Turbo Signal Recovery for Nonlinear Measurements and Orthogonal Sensing Matrices
• Statistical-mechanical analysis of pre-training and fine tuning in deep learning
• Online compressed sensing
• Analyticity of the energy in an Ising spin glass with correlated disorder
• Expectation propagation on the diluted Bayesian classifier