基于随机森林的建筑物场景微蜂窝模型的构建  

CONSTRUCTION OF BUILDING SCENE MICROCELLULAR MODEL BASED ON RANDOM FOREST

作　　者：陈明发 严天峰[1,2,3] Chen Mingfa;Yan Tianfeng(School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,Gansu,China;Radio Monitoring and Technology Center of Positioning Industry of Gansu Province,Lanzhou 730070,Gansu,China;High-precision Positioning Technology Compass Engineering Laboratory of Gansu Province,Lanzhou 730070,Gansu,China)

机构地区：[1]兰州交通大学电子与信息工程学院,甘肃兰州730070 [2]甘肃省无线电监测及定位行业技术中心,甘肃兰州730070 [3]甘肃省高精度北斗定位技术工程实验室,甘肃兰州730070

出　　处：《计算机应用与软件》2020年第10期119-124,共6页Computer Applications and Software

基　　金：甘肃省高等学校创新团队项目(2017C-09);中国铁路总公司科技研究开发计划重大课题(2017XO13-A)。

摘　　要：在市区微蜂窝中,信号的传播由于建筑物阻挡衰减很大。传统的微蜂窝模型在测试地区实际应用时存在很大误差的问题,对此提出改进市区微蜂窝传播模型。通过测试计算路径损耗斜率,提升了自由空间损耗模型精度,以改进算法分析建筑物损耗为基础进行建模。实验结果表明,本地区的路径损耗斜率为23.51 dB/dec,改进路径损耗斜率的自由空间损耗模型与传统模型相比,精度提升了1.5 dB;基于随机森林构建建筑物损耗模型明显优于其他算法构建的建筑物损耗模型。In urban microcells,signal propagation is greatly attenuated due to building block.Because the traditional microcellular model has a lot of errors in the practical application of the test area,the paper proposes to improve the urban microcellular propagation model.By testing and calculating the slope of path loss,the accuracy of free space loss model is improved,and the model is based on the analysis of building loss by the improved algorithm.The experimental results show that the path loss slope in this region is 23.51 dB/dec,and the accuracy of the improved free space loss model is 1.5 dB higher than that of the traditional free space loss model.The building loss model based on random forest is obviously superior to building loss model based on other algorithms.

关 键 词：微蜂窝 随机森林 路径损耗 误差 

分 类 号：TN011[电子电信—物理电子学]