GNSS-5G Hybrid Positioning Based on Multi-Rate Measurements Fusion and Proactive Measurement Uncertainty Prediction

Abstract

The fifth-generation (5G) network can be used jointly with a global navigation satellite system (GNSS) to help relieve the problem of having few visible satellites in urban environments and achieve highly accurate localization. The 5G base stations (BSs) can provide range and angle measurements at a much higher rate than GNSS. However, this benefit has not been fully employed by existing GNSS-5G hybrid methods. In addition, the near–far effect of 5G BS causes a large dynamic range of the measurement noise uncertainty. This challenge has not been well dealt with by the existing adaptive noise estimation methods, resulting in the decline of positioning accuracy and system convergence. This work proposes a multiple-rate adaptive Kalman filter (MRAKF) for GNSS-5G hybrid positioning with a hybrid sequential fusing scheme. It supports the fusion of GNSS and 5G measurements at different data rates to better employ the high-rate 5G measurements. A distance-dependent model is derived to quantitatively characterize the effect of BS-user equipment (UE) distance over the measurement noise covariance, leading to a proactive measurement uncertainty prediction algorithm to adaptively adjust the observation noise covariance matrix. Furthermore, a switchover strategy is proposed to switch its positioning mode to avoid the impact of variation of available line of sight (LOS) BSs. Both simulations and a real driving test were carried out. Results show that the proposed MRAKF method significantly improves the positioning accuracy compared with the other four adaptive noise estimation methods and the standard EKF. It requires the lowest computation complexity among all adaptive methods in this work.