5G + Beidou fusion positioning method based on neural network

The invention relates to the technical field of satellite navigation, in particular to a 5G + Beidou fusion positioning method based on a neural network, and the method comprises the steps: S1, selecting a reference point of a to-be-positioned region of UE, obtaining TOA data of a BDS signal and RSS data of a 5G signal, building a fingerprint database, and training a system model; s2, a BDS signal is captured and tracked, the TOA value of the position where UE is located is obtained, then filtering is carried out through a Kalman filter, and a part of noise in TOA is eliminated; s3, the RSS value of the position where the UE is located is measured, and coarse estimation of the position is achieved through an RSS fingerprint neural network; and S4, fusing the TOA of the BDS signal and the RSS of the 5G signal, and obtaining the accurate position of the UE through a data fusion positioning network. 本发明涉及卫星导航技术领域，尤其是涉及一种基于神经网络的5G+北斗融合定位方法，包括：S1：选取UE待定位区域参考点，获取BDS信号的TOA数据以及5G信号的RSS数据，建立指纹数据库，训练系统模型；S2：捕获并跟踪BDS信号，