C-Cube: Rethinking Distributed Beamforming for Concurrent Charging in Backscatter Networks

Recent innovations significantly advance the communication capabilities of backscatter networks, while energy harvesting remains the bottleneck of these backscatters. To tackle this problem, existing studies explore distributed beamforming techniques to beam enhanced energy for a battery-free tag. However, they cannot concurrently transfer high-density power to multiple tags since the system is unable to estimate accurate channel state information (CSI). In this paper, we argue that by deploying a lightweight beamforming helper, we can push the limit of CSI estimation with negligible overhead on the tag. The big gap between the wireless charging threshold and the CSI estimation sensitivity can provide stable estimation result of CSI for the charging system even the signal experiences double fading. On this basis, we propose C-Cube, a concurrent charging system for backscatter networks. C-Cube employs distributed beamforming with the help of CSI to indicate the phase alignment process. To bring it into practice, we design a cold start scheme to activate tags by transmitting a distributed orthogonal frequency (DOF) signal. Moreover, to ensure all tags' CSIs have been successfully received, a spatial-based CSI classifier is explored in C-Cube. We further align the beamforming phase to optimize a novel metric called relative power, instead of the received power metric. The new metric can eliminate the impact of the unknown tag-to-helper channel. We implement our system with USRP/GNURadio platform and customized tags in a real environment. C-Cube can achieve the same coverage as the state-of-the-art and reduce the charge time by 44.1% and 30.2%.