基于聚类分析的反向射线跟踪算法

摘要

现有无线定位引入射线跟踪算法环境建模过于理想,不适合实际应用.为此,提出了一种基于聚类分析的反向射线跟踪算法.为贴近射线的实际传播环境,构建了建筑物表面为准镜面的二维传播环境模型;根据多径定位信号传播路径的特征定义了位置指向性路径(P-路径)和区域指向性路径(A-路径),算法根据多径定位信号的到达时间和到达角度的估计值,利用分支跟踪搜索算法找出其可行路径,再用聚类分析方法实现对位置指向性路径的识别和传播路径跟踪,并采用分段加权的思想完成对区域指向性路径的最大可识别路径的跟踪.仿真结果验证了该算法的合理性,与基于绕射判决门限的反向射线跟踪算法相比,提出的算法能显著提高对P-路径和A-路径随TOA和AOA估计误差变化的跟踪正确率.%The environmental modeling of existing ray tracing algorithms is too ideal,which can not fit the practical application.To solve this problem,this paper proposed a reverse ray tracing algorithm based on clustering analysis.Firstly,in order to fit the actual propagation environment better,it constructed a 2D modeling which considered the building surface as quasimirror.Secondly,according to the propagation characteristics of the multiple-path location signal,this paper defin the position-orientating path(P-path) and the area-orientating path(A-path),and found the feasible paths by using the branch tracking and searching method according to the TOA and AOA estimates of the location signal.It distinguished and traced the Ppaths by using the clustering analysis method,then traced the maximum recognizable paths for A-paths according to the idea of subsection weighting.In the end,the simulation results show that the proposed reverse ray tracking algorithm is reasonable,and compared with methods based on diffraction threshold,the proposed algorithm significantly improves the accuracy of the ray tracking for both P-path and A-path according to different TOA and AOA estimates error.