High Precision Positioning using Multi-cell Massive MIMO system for 5G and beyond

Abstract

The large number of antennas at the base stations (BSs), the reference signal transmission over a wide bandwidth, and dense network deployment can be used to achieve unprecedented position accuracy in 5G and beyond networks. This work proposes a position estimation method using a full dimension massive multiple-input multiple-output (MaMIMO) system in cellular networks. The user equipment (UE) estimates the time of flight/arrival (ToA) from multiple BSs using estimation of signal parameters via rotational invariant technique (ESPRIT). A novel method that selects a subset of the accurate ToA measurements and uses it to estimate the position of the target UE based on multilateration is proposed. The simulation results demonstrate that there is a significant improvement in positioning accuracy over the conventional methods, and the proposed method closely matches the genie-aided algorithm for small subsets, and outperforms it for large subsets. The proposed method achieves 20 cm positioning accuracy for 90% of the UEs in indoor factory (InF) scenarios using practical 3GPP channel models. It also satisfies the ITU requirement for Industry 4.0 and requires a low transmission and measurement overhead.