一种UDN天线受限条件下的定位方法  

UDN positioning method in the case of constrained antennas

作　　者：谢亚琴[1] 郑迪 谷天园 XIE Yaqin;ZHENG Di;GU Tianyuan(School of Electronic and Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China)

机构地区：[1]南京信息工程大学电子与信息工程学院,南京210044

出　　处：《中国惯性技术学报》2023年第10期976-985,共10页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金(62001238)。

摘　　要：在超密集网络(UDN)环境中,当部分阵列天线单元受损且UDN服务基站天线单元数量较少时,信道状态信息(CSI)会产生随机丢失现象,精确估计用户的三维位置信息成为一个巨大的挑战。针对上述问题,提出了一种UDN天线受限条件下的定位方法。首先,将CSI张量按mode-1展开,并检测其信息缺失位置。然后,基于奇异值阈值(SVT)算法对缺失的CSI进行补全,并将补全后的CSI矩阵重构为张量。最后,基于Tucker分解方法估计信号的时延、方位角和仰角,利用基站与用户之间的几何关系得到用户的三维位置坐标。仿真结果表明,所提方法在UDN天线数量受限且部分受损的条件下表现出一定优势,当信噪比为-20 dB,天线数量为16,总CSI随机丢失率为2/16,多径数为3时,所提方法在实时场景下位置均方根误差为3.92 m,且相较于Tucker分解方法,定位精度提高了25.33%。In the ultra-dense network(UDN)environment,when some array antenna units are damaged and the number of UDN service base station antenna units is small,the channel state information(CSI)will be randomly lost,which becomes a huge challenge to accurately estimate the user's 3D position information.Aiming at the above mentioned problem,a UDN positioning method in the case of constrained antennas is proposed.Firstly,the CSI tensor is expanded by mode-1 and the location of the information missed is detected.Then,the missing CSI is completed based on singular value threshold(SVT)algorithm,and the completed CSI matrix is reconstructed into a tensor.Finally,the time delay,azimuth and elevation of the signal are all estimated based on the Tucker decomposition method,and the three-dimensional position of the user is obtained by using the geometric relationship between the base station and the unknown user.The simulation results show that the proposed method has certain advantages under the condition that the number of UDN antennas is limited and partially damaged.When the signal-to-noise ratio(SNR)is-20 dB,the number of antennas is 16,the total CSI random loss rate is 2/16,and the multipath number is 3,the position root-mean-square error of the proposed method is 3.92 m in real-time scenario.Compared with the Tucker decomposition method,the positioning accuracy is improved by 25.33%.

关 键 词：天线单元受损 矩阵补全 Tucker分解 三维定位 

分 类 号：TN92[电子电信—通信与信息系统]