Joint reconstruction algorithms for one-bit distributed compressed sensing

Abstract

Distributed compressed sensing (DCS), exploiting the correlation among multiple signals, enjoys the advantage of reduced number of measurements. This paper considers a type of joint sparsity model in DCS, where each signal contains a common component and an innovation component. In order to reduce the transmission cost, the measurements are derived as the sign information of the compressed samples by using one-bit quantization. We study such CS operation, and propose two joint reconstruction algorithms by iteratively deriving the sign information of each component. Simulation results show that the proposed algorithms can recover the signals efficiently.