SWALLOW:一个稳定的WLAN区域定位系统  

SWALLOW:A Stable WLAN Area Localization System

作　　者：易飞滔[1,2] 黄刘生[1,2] 刘刚[1,2] 徐宏力 

机构地区：[1]中国科学技术大学计算机科学与技术学院,合肥230027 [2]中国科学技术大学苏州研究院,江苏苏州215123

出　　处：《小型微型计算机系统》2013年第6期1237-1241,共5页Journal of Chinese Computer Systems

基　　金：国家科技重大专项项目(2011ZX0300500404)资助;国家"九七三"重点基础研究发展计划项目(2011CB302905)资助;国家自然科学基金项目(61170058)资助;广东省中国科学院全面战略合作项目(2011A090100012);苏州市工业技术专项项目(ZXG201041)资助

摘　　要：介绍WLAN定位系统SWALLOW的设计与实现.本系统对WLAN中携带定位卡的人员或设备实施定位,考虑到定位卡的成本与能耗,本系统采用接入点而非定位卡收集RSSI值.而各接入点对定位卡同一信号的RSSI测值往往存在缺失和不同步,因此传统的基于点的RSSI特征比对算法稳定性不好.本系统使用了一个新的基于区域的定位算法,该算法在最小信号距离比对的初步结果上,通过在生成树上寻找最近公共祖先的方法实现平滑.分析和实验表明,该算法能将静止定位稳定度从50%提高到95%,并保持90%以上的移动定位准确度和10s以下的延时.目前,SWALLOW系统已应用到资产管理系统中.Presented the design and implementation of the WLAN-based localization system called SWALLOW. The system compute the location of a person or an equipment which carries a localization card, Given the cost and energy consumption of the localization card, it uses access points instead of localization cards to collect RSSI values. However, the RSSI values collected by access points are prone to transmission failure and unsynchronized, which results in the unstable localization by the traditional point-based finger- print algorithms. The system uses a new area-based localization algorithm, which gets a preliminary result by comparing distance in signal space, then gets a smooth result by finding lowest common ancestors in spanning tree. Analysis and experiments show that the static localization stability is raised to 95% from 50% by the algorithm, the mobile localization accuracy exceeds 90%, and the local- ization delay is less than 10s. The SWALLOW system has been applied to an asset management system.

关 键 词：WLAN 区域定位 生成树 最近公共祖先 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]