Physical-Layer Abstraction for Hybrid GNSS and 5G Positioning Evaluations

Abstract

Hybridization of Global Navigation Satellite Systems (GNSS) and fifth generation (5G) cellular positioning is foreseen as a key solution to fulfill high-accuracy positioning requirements in future use cases, such as autonomous vehicles. The evaluation of the hybrid positioning capabilities implies the physical-layer simulation of observables from both GNSS and 5G technologies. In order to ease the complexity of the resulting system-level simulations, a physical-layer abstraction of GNSS and 5G ranging observables is here proposed. The abstraction of GNSS ranging observables is based on a Gaussian-distributed model of the errors sources, while the abstraction of 5G ranging observables is based on the interpolation of the cumulative density function (CDF) of the ranging errors for certain propagation conditions and signal-to-noise (SNR) levels. Thanks to the exploitation of the proposed physical-layer abstraction, low-complexity system- level simulations are performed to assess the positioning capabilities of GNSS and 5G downlink time-difference of arrival (DL-TDoA) in urban macro-cell (UMa) environments. The simulation results indicate the need to adopt hybrid solutions based on multiple GNSS constellations and 5G DL-TDoA with 100-MHz bandwidth, in order to ensure a horizontal positioning accuracy below 5 m for 95% of cases in outdoor urban environments.