面向多重应用的高鲁棒被动式定位模型研究  

Research on High Robust Device-Free Localization Model for Multiple Applications

作　　者：陈晓江[1,2] 陈丽丽 李博航 汤战勇 谢彬彬[1] 王薇 王安文[1,2] 房鼎益[1,2] CHEN Xiao-Jiang;CHEN Li-Li;LI Bo-Hang;TANG Zhan-Yong;XIE Bin-Bin;WANG Wei;WANG An-Wen;FANG Ding-Yi(School of Information Science and Technology, Northwest University, Xi’an 710127;Shaanxi International Joint Research Center for Battery-Free Internet of Things, Xi’an 710127)

机构地区：[1]西北大学信息科学与技术学院,西安710127 [2]陕西省无源物联网国际联合研究中心,西安710127

出　　处：《计算机学报》2019年第2期403-414,共12页Chinese Journal of Computers

基　　金：国家自然科学基金(61572402;61672428;61772422);陕西省创新团队(2018TD-026)资助~~

摘　　要：基于无线信号的被动式目标定位技术由于具备无需目标携带设备、便于实施、对视线和监测视角要求不高等优势,因此在老人监护、家居自动化以及野生动物监测等应用领域具有巨大的潜力.传统的被动式定位方法为保证定位精度,要么需要大规模训练学习,导致勘测代价大;要么只针对于某些专用设备或特殊应用,导致普适性差.对此,文中提出一个面向多重应用的高鲁棒被动式定位模型MaLoc:无需大规模训练学习,并且可在廉价商用设备上实现.MaLoc的主要特点在于:(1)通过获取接收端从不同天线接收信号的稳定相位差,结合参考阵列天线的位置来实现基于到达角度的目标定位,保证了商用设备下的定位精度;(2)为了实现高鲁棒的到目标反射信号到达角度估计,MaLoc首先利用静止干扰源的角度误差补偿来缓减信号到达角度随时间产生的偏移;其次利用不同种类反射信号之间的相干与非相干性,对环境中移动干扰源和静止干扰源反射信号的到达角度进行消除.如此进一步提升了定位的鲁棒性和精度.最后通过实际环境下的多种实验验证了MaLoc的有效性和鲁棒性.Device-free localization technology based on wireless signal has become a burgeoning research direction in wireless sensor network.It enables extensive applications such as elderly care,home automation and wildlife monitoring.Due to the key advantages that neither requires user carry any devices nor needs special light or angle conditions.During the several past decades,remarkable successes have been made in this field.However,to our best knowledge,traditional device-free localization methods drove by fingerprint and model both experience various limitations,for achieving high localization accuracy.Specifically,the fingerprint based approaches require large scale training that results in great human effort;the model based approaches aim at dedicate devices and special applications which lead to poor universality.In this paper,we present MaLoc,a robust device-free localization model for multiple applications,which does not require extensive prior learning,and can work effectively under both cheap commercial devices and dedicate devices.We utilize the intuition that whenever the object blocks a wireless signal coming from a certain direction,the direction of arrival representing the corresponding direction will experience a sharp drop in signal strength.Thus,the interface can identify the object’s location by leveraging the direction of arrival values of incoming signals at the receiver.The main innovation of this paper includes two aspects.(1)To mitigate the error ofr phase measurements collected by commercial cheap WiFi device we make specific linear transformation to the raw phase measurement to obtain the stable available phase eigenvalues,which can be used to calculate stable phase difference.By getting the stable phase difference between different receipt signals from array antenna from receiving terminals,meanwhile considering the known location of reference antenna,we can determine the target location basing on arrival angle.The processing of the unstable phase guarantees the localization accuracy of syst

关 键 词：被动式定位 高鲁棒 信号到达角度 角度误差补偿 相干 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]