Capo: Calibrating Device-to-Device Positioning with a Collaborative Network

Kao Wan, Zhaoxi Wu, Qian Cao, Xiaotao Zheng, Ziwei Li, Tong Li

Abstract: Device-to-device relative positioning has been widely applied in modern Web-based systems such as COVID-19 mobile contact tracing, seamless access systems, mobile interactive gaming, and mobile e-Commerce. The legacy absolute positioning technologies are not suitable for device-to-device positioning attributed to their mobility and heterogeneity of devices. In this paper, we focus on the heterogeneity problem and propose Capo, the first calibration algorithm that enables the interaction among devices with different communication modes for relative positioning in heterogeneous systems. Capo optimizes the ranging results of low-precision devices in a collaborative network based on the ranging data from high-precision devices. The evaluation shows that Capo can significantly improve up to 26.56% of the positioning accuracy of the heterogeneous systems. Real use case study on COVID-19 contact tracing further shows that Capo significantly improves the accuracy of exposure notifications.