Positioning and Dynamic Tracking under 3GPP Cellular Network on GNURadio Testbed

This paper addresses high precision 5G positioning and dynamic tracking, which has risen increasingly attention from industry and academia recently. The recent works show that the position of user equipment (UE) could be estimated based on reference signals (e.g. PRS) in 5G on real testbed. However, these works failed to exploit the positional information behind the consecutive nature of users' moving, which could enhance the positioning accuracy by exploiting dynamic tracking architectures. In this paper, we extract the angle of arrival (AoA) and time of arrival (ToA) under 3GPP physical layer protocol, and then introduce a dynamic tracking method to improve the accuracy of AoA and ToA estimation. The performance of the proposed method was evaluated on GNURadio. The results show that the AoA and ToA could be precisely extracted in the real propagation channel on the premise of guaranteeing the communication performance. And in the case of the same bandwidth and number of antennas, compared with traditional base station (BS) tracking method, our dynamic tracking could reduce tracking error by 35.11% during the 115s tracking process.