摘要
随着基于位置服务(Location Based Services,LBS)的发展与智能移动设备的普及,室内定位算法与系统受到了广泛研究与关注。为提高室内定位精度、增强系统鲁棒性,提出了基于多边限定的fingerprint定位方法。基于Wi-Fi RSSI(Received Signal Strength Indication)信号处理建立离线fingerprint数据库;通过对拟合距离-RSSI函数分析,提出了多边限定的方法确定一个最佳参考点(Reference Point,RP)集合,缩小在线定位阶段的搜索范围。在此基础上,再利用fingerprint定位方法进行定位。此外,实现了基于提出方法的室内定位系统原型用于算法性能评估。通过大量真实场景实验分析、验证了相较于传统fingerprint方法,基于多边限定的fingerprint定位方法能有效提高室内定位精度,增强系统鲁棒性。
With the development of the Location Based Service(LBS)and the prevalence of smart mobile devices,indoor localization techniques have aroused great attention and been widely studied. This work aims at proposing a multilateralconfined method for fingerprint-based localization to improve the localization accuracy and enhance the system robustness.First, it establishes the offline fingerprint database based on Wi-Fi Received Signal Strength Indication(RSSI)processing.Then, based on the analysis of distance-RSSI function fitting, it proposes a multilateral-confined method to determine a set of Reference Points(RP), which is used to narrow down the search space during the online localization phase. Then,the fingerprint-based method is adopted to estimate the real-time location. Besides, for performance evaluation, it implements the prototype of the localization system as well as the proposed solution. A comprehensive real-world experiment is conducted and the results demonstrate that the proposed solution is able to effectively improve the localization accuracy and enhance system robustness compared with conventional fingerprint-based localization techniques.
作者
夏宇声
刘凯
张浩
金飞宇
郑林江
XIA Yusheng;LIU Kai;ZHANG Hao;JIN Feiyu;ZHENG Linjiang(College of Computer Science, Chongqing University, Chongqing 400044, China)
出处
《计算机工程与应用》
CSCD
北大核心
2019年第11期250-256,共7页
Computer Engineering and Applications
基金
国家自然科学基金(No.61572088)
国家高技术研究发展计划(863)(No.2015AA015308)
重庆市自然科学基金(No.cstc2017jcyjAX0026)
