Joint Localization and Signal Detection for Ambient Backscatter Communication Systems

Ambient backscatter communication (AmBC) has emerged as a promising technology for passive Internet of Things (IoT), which enables backscatter devices (BDs) to transmit information over ambient radio frequency (RF) signals. This paper investigates joint localization and signal detection for an AmBC system. The BDs modulate information over ambient orthogonal frequency division multiplexing (OFDM) signals, and the receiver realizes joint localization and signal detection for the BDs. A two-stage receiver algorithm is proposed to realize the joint localization and signal detection. In the first stage, the receiver estimates the delays and angles-of-arrival (AoAs) of the BDs through the orthogonal matching pursuit (OMP) based algorithm. Then attenuation coefficients of the backscatter link are estimated through the least squares (LS) method, based on which the information symbols transmitted by the BDs can be detected in the second stage. Further, to address the discretization error in the estimation of AoAs and delays through the OMP algorithm, we propose an sparse Bayesian learning (SBL) based algorithm to achieve off-grid estimation. Finally, the Cramér-Rao lower bound (CRLB) is derived to evaluate the performance of the algorithms. Simulation results have verified the effectiveness of the system model and the superiority of the SBL-based algorithm, and it is shown that larger bandwidth and antenna arrays are beneficial to improving the estimation accuracy.