城市峡谷中基于GPS双极化天线的定位方法

Positioning method based on GPS dual-polarized antenna in urban canyon environment

针对在城市峡谷环境中,卫星信号易受建筑遮挡或反射产生多径效应,导致定位精度降低的问题,提出基于全球定位系统(GPS)双极化天线的多径/非视距信号剔除和修正定位方法:将左、右旋圆极化天线输出的载噪比与卫星高度角、伪距残差作为输入特征,使用梯度提升决策树对伪距误差进行拟合;并使用训练得到的规则对伪距误差进行预测,通过将该预测值的绝对值与阈值进行比较来判断信号接收类型;然后剔除多径/非视距信号或对其伪距进行修正。实验结果表明,所提出的方法的分类准确率达到78%,其2维(2D)和3维(3D)定位的均方根误差分别比传统定位方法提高43.07%和41.76%,能有效提高城市峡谷内的定位精度。

Aiming at the problem that it is liable to multipath effects due to the blockage and reflection of buildings under urban canyon environments for satellite signals,resulting in reducing the positioning accuracy,the paper proposed a positioning method of elimination and correction of multipath/non-line-of-sight(NLOS)signals based on GPS dual-polarized antenna:the carrier-to-noise ratio of the left and right circularly polarized antennas together with satellite elevation angle and pseudorange residual were used as features to fit the pseudorange errors by using the gradient boosting decision tree algorithm;and the extracted rules from the fitting were used to predict the pseudorange errors,then the signal receiving types were determined by comparing between the absolute and threshold values of the predicted data;finally the multipath/NLOS signals were excluded or corrected according to their sizes of pseudorange errors.Experimental result showd that the proposed method would reach the classification accuracy of 78%,and the root mean square error(RMSE)of 2D and 3D positioning would be 43.07%and 41.76%respectively higher than those of traditional methods,indicating that the suggested algorithm could efficiently improve the positioning accuracy in urban canyons.