作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 移动互联与通信技术 • 上一篇    下一篇

基于WiFi 指纹的室内轨迹定位模型

蔡文学1,邱珠成1,黄晓宇1,萧超武1,陈 康2   

  1. (1. 华南理工大学经济与贸易学院,广州510006; 2. 中国电信股份有限公司广东研究院,广州510630)
  • 收稿日期:2014-06-10 出版日期:2015-06-15 发布日期:2015-06-15
  • 作者简介:蔡文学(1968 - ),男,教授,主研方向:物联网,智能交通;邱珠成,硕士研究生;黄晓宇(通讯作者),讲师、博士;萧超武,硕士研 究生;陈 康,硕士。
  • 基金资助:

    国家“863”计划基金资助项目“业务受控的精密定位信息移动广播系统技术”(2012AA12A203);2011 年粤港关键领域重点突破基金资助项目“乳制品质量安全全生命周期智能溯源平台建设及应用”(2011A011303002)。

Indoor Track Positioning Model Based on WiFi Fingerprint

CAI Wenxue  1,QIU Zhucheng  1,HUANG Xiaoyu  1,XIAO Chaowu  1,CHEN Kang  2   

  1. (1. School of Economics and Commerce,South China University of Technology,Guangzhou 510006,China; 2. Academy of Guangdong,China Telecom Co. ,Ltd. ,Guangzhou 510630,China)
  • Received:2014-06-10 Online:2015-06-15 Published:2015-06-15

摘要:

基于WiFi 指纹的无线定位根据目标接收到的无线信号强度来反推其所处位置的定位模型,在室内定位研究中被广泛应用。当前对指纹定位算法的研究主要集中在单点定位,但在实际的应用中经常需要追踪目标在空间上的运动轨迹。为此,在单点定位模型的基础上,提出一种基于核函数的隐马尔科夫链模型,通过高斯核函数计算指纹的似然概率,定义位置点之间的转移概率,同时通过限制转移位置点的搜索范围提高算法效率,并应用隐马尔科夫链模型对移动轨迹进行定位。实验结果表明,该算法在定位准确率和定位误差方面性能均优于对比算法。

关键词: 室内定位, 指纹定位, 轨迹定位, 隐马尔科夫模型, 核函数

Abstract:

As one of the most promising indoor positioning technologies,the WiFi-based fingerprinting model has attracted much research attention in recent years. However,most of the existing studies mainly focus on the single point positioning method,while in the actual applications,it often needs to track the object trajectory. Hence, in this paper, the Hidden Markov Model ( HMM ) based on kernel function is presented, which is on the basis of the single-point positioning model. The Gaussian kernel function is used to calculate the likelihood between fingerprints. The transition probability is defined and the search range is limited by a threshold value, which can improve the efficiency of the algorithm. HMM is applied to the track positioning. Experimental result shows that the proposed algorithm outperforms the benchmark algorithms significantly on both of the precision and distance measures.

Key words: indoor positioning, fingerprint positioning, track positioning, Hidden Markov Model(HMM), kernel function

中图分类号: