摘要
针对城市峡谷以及室内环境,卫星信号强度大幅降低,导致定位精度不能满足需求以及在室内环境下需要联合多个射频单元才能进行定位的问题,提出一种基于信道图谱的5G+北斗融合定位方法,在室内环境下,将位置信息与信道特征直接进行关联,将信道特征以图像的方式进行展现,将定位的问题转化为图像识别的问题,采用3D卷积神经网络将信道图像和位置信息关联起来,在室外环境下将5G定位与北斗定位进行融合,采用最小二乘拟合算法来处理测量数据,建立北斗+5G联合定位模型,实现联合定位。
In urban canyons and indoor environments,the strength of satellite signals will be greatly reduced,resulting in insufficient po-sitioning accuracy to meet the needs and the need to combine multiple radio frequency units for positioning in indoor environ-ments.A 5G+Beidou fusion positioning based on channel map is proposed.method,in the indoor environment,the location information is directly associated with the channel features,the channel features are displayed in the form of images,and the problem of positioning is transformed into the problem of image recognition.Integrate 5G positioning and Beidou positioning in outdoor environment,use least squares fitting algorithm to process measurement data,establish Beidou and 5G joint position-ing model,and realize joint positioning.
作者
张千坤
陈任翔
曾炜
钟志刚
刘西西
张宸宾
Zhang Qiankun;Chen Renxiang;Zeng Wei;Zhong Zhigang;Liu Xixi;Zhang Chenbin(China Information Technology De-signing&Consulting Institute Co.,Ltd.,Beijing 100048,China;China Unicom Jiangxi Branch,Nanchang 330000,China)
出处
《邮电设计技术》
2023年第11期33-39,共7页
Designing Techniques of Posts and Telecommunications
关键词
北斗
卷积神经网络
信道图谱
图像识别
室内定位
BDS
Convolutional neural network
Channel map
Image identification
Indoor position
