基于传播图论的多链路信道仿真及其在TDOA定位系统性能评估中的应用  

Multi-link channel simulation based on propagation graph theory and application in performance evaluation of TDOA positioning system

作　　者：段嘉伟 薛冰岩 尹学锋[1] DUAN Jiawei;XUE Bingyan;YIN Xuefeng(College of Electronics and Information Technology,Tongji University,Shanghai 201804,China)

机构地区：[1]同济大学电子与信息工程学院,上海201804

出　　处：《电波科学学报》2021年第3期413-421,共9页Chinese Journal of Radio Science

摘　　要：为了解决在真实环境下到达时间差(time difference of arrival,TDOA)定位性能评估所面临的测量成本高、不可控干扰因素较多等问题,提出了一种利用传播图论多链路信道仿真来评估TDOA定位系统性能的方法.即采用传播图论的方法生成不同的信道冲激响应,并通过软件无线电装置产生经信道畸变的射频信号,作为多个TDOA传感器的输入,形成一种新型的TDOA定位检测方法.首次将复杂外场采用射频仿真的方式,针对城市环境中的TDOA系统构建4个测试场景,对TDOA系统的定位精度进行检验.仿真结果表明,被测的多链路定位系统在信号带宽不小于200 kHz条件下表现出良好且稳定的性能.此外,本文提出的嵌入衍射机制的图论建模方法,相比传统图论更为有效地提升了TDOA定位误差的仿真准确度.In order to solve the problems of high measurement cost and many uncontrollable interference factors in the accuracy evaluation of time difference of arrival(TDOA)localization method by measurement,a method of evaluating the accuracy of a TDOA localization system based on multi-link channel propagation graph theory is proposed.In this paper,we use the propagation graph theory to generate different channel impulse responses,and ultilize software-defined radio devices to generate channel-distorted radio frequency signals as inputs to multiple TDOA sensors and construct a novel TDOA localization evaluation method.In order to verify the accuracy of the TDOA technology effectively,we simulate the method in 4 typical scenarios.It is the first time that a complex scenario is simulated using multi-channel radio frequency signal to test the localization accuracy of the TDOA system.The simulation result shows that when the signal bandwidth is not smaller than 200 kHz,the multi-link localization system based on graph modeling shows good and stable performance.Moreover,graph modeling method with diffraction model embedded enhances the accuracy of performance simulation for the TDOA system,particularly in the non-line-of-sight scenarios.

关 键 词：信道建模 传播图论 衍射模型 多链路到达时间差 定位算法 大数据 

分 类 号：TN929.53[电子电信—通信与信息系统]