5G Indoor Positioning Performance Evaluation Based on Semi-Deterministic Channel Model

The 5th generation mobile communication system is considered a promising technical solution for the universal indoor positioning problem due to its high bandwidth, dense base stations, and large antenna scale. Indoor scenes are characterised by a deterministic propagation environment. The positioning performance evaluation of statistical channel models in specific indoor scenes is subject to non-negligible deviations, and deterministic channel modelling therefore requires appropriately detailed propagation environment information. To this end, this paper improves a semi-deterministic channel modelling method. The method is based upon the statistical characteristics of the scene data and uses a ray-tracing technique in accordance with key propagation environment information. Direct radiation, reflection, and scattering mechanisms of multipath signals are considered as part of the model. Additionally, it is capable of mining the inherent propagation characteristics of wireless signals, and evaluating the positioning performance in specific indoor scenarios. The channel model allows effortless access to a location fingerprint database based on 5G wireless signals in specific indoor scenarios. Through controlled experiments, we evaluated the impact of various factors on the positioning performance of the fingerprint database. The experimental results provide us with a direction to improve positioning performance.