FISCP: fine-grained device-free positioning system for multiple targets working in sparse deployments
FISCP: fine-grained device-free positioning system for multiple targets working in sparse deployments
复制标题
DOI：
10.1007/s11276-016-1253-8
复制
发表时间：
2016-03
影响因子：
3
通讯作者：
Binbin Xie;Dingyi Fang;Tianzhang Xing;Lichao Zhang;Xiaojiang Chen;Zhanyong Tang;Anwen Wang
中科院分区：
文献类型：
--
作者：
Binbin Xie;Dingyi Fang;Tianzhang Xing;Lichao Zhang;Xiaojiang Chen;Zhanyong Tang;Anwen Wang
无设备定位技术一直以来都在反入侵应用中扮演着重要的角色。然而，目前的多目标定位解决方案主要使用不经济的设备，这对于大规模场景来说不具有成本效益。此外，它们只考虑了被目标扭曲的视线（LoS）路径信号，因此当面临多径效应和非视线（NLoS）时，无法准确定位，这在现实世界的部署中是典型的。在本文中，我们提出了FISCP，一个细粒度的无设备定位系统的多个目标工作在稀疏部署。采用RFID无源标签，这是比其他设备的定位便宜得多。同时，FISCP利用目标在多径环境中产生的动态失真，将其作为指纹进行定位，而以往的定位方法忽略了多径效应，甚至认为多径是有害的。我们使用商业现成的产品，包括RFID系统和全向天线，制作了一个原型系统的FISCP，并开发了一个软件程序的RFID系统。所有的实验中进行的部署，其中每对标签之间的距离间隔为1.2米，和部署是稀疏相对于无源RFID系统的短通信范围（从几米到几十米）。实验结果表明，FISCP在多目标定位中是有效的，平均定位误差为0.33 m。
Device-free localization has long been playing a key role in the anti-intrusion applications. However, current multi-target localization solutions mainly use uneconomical equipment, which are not cost-efficient for large-scale scenarios . Moreover, they only consider the line-of-sight (LoS) path signals distorted by the targets, hence can’t exactly pinpoint the locations when faced with multipath effects and non-line-of-sight (NLoS), which are typical in real-world deployments. In this paper, we propose FISCP, a fine-grained device-free positioning system for multiple targets working in sparse deployments. The RFID passive tags are employed, which are much cheaper than other devices for localization. Meanwhile, unlike past approaches, which ignore the multipath effects or even take multipath as detrimental, FISCP exploits the dynamic distortion in multipath caused by the targets and considers the distortion as fingerprints for localization. We make a prototype system for FISCP using the commercial off-the-shelf products, including RFID systems and omnidirectional antennas, and develop a software program for the RFID systems. All the experiments are conducted in the deployments where the distance interval between each pair of tags is 1.2 m, and the deployments are sparse with respect to the short communication range of passive RFID systems (from a few meters up to tens of meters). The results of our experiments demonstrate that FISCP is effective in multi-target localization with low localization errors of 0.33 m in average.