计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (1): 29-34.DOI: 10.3778/j.issn.1002-8331.1809-0143

• 热点与综述 • 上一篇    下一篇

基于深度学习的Wi-Fi与iBeacon融合的室内定位方法

薛  伟,陈  璟,张  熠   

  1. 1.江南大学 物联网工程学院,江苏 无锡 214122
    2.江南大学 物联网技术应用教育部工程研究中心,江苏 无锡 214122
  • 出版日期:2019-01-01 发布日期:2019-01-07

Indoor Localization Based on Deep Learning Using Wi-Fi and iBeacon

XUE Wei, CHEN Jing, ZHANG Yi   

  1. 1.School of Internet of Thing Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
    2.Engineering Research Center of Internet of Things Technology Applications, Ministry of Education, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Online:2019-01-01 Published:2019-01-07

摘要:

针对传统室内定位指纹法存在定位精度低、容易受到环境影响的问题,提出了一种基于深度学习的Wi-Fi与iBeacon融合的室内定位方法。离线阶段在参考点处采集各个AP和iBeacon的信号强度,使
用这些信号强度数据对堆叠自动编码机进行训练并从大量带有噪声的信号强度样本中提取特征,构建位置指纹数据库;在线定位阶段,使用堆叠自动编码机获得待测点信号强度特征并与位置指纹数据库中信
号强度特征进行匹配,通过近邻算法估计待测点位置。实验结果表明,基于堆叠自动编码机的室内定位算法具有更高的定位精度。

关键词: 室内定位, 深度学习, 堆叠自动编码机, 近邻算法, iBeacon, Wi-Fi

Abstract:

Aiming at the problem that the traditional indoor fingerprint localization algorithm has low positioning accuracy and is easily affected by the environment, an indoor
localization algorithm based on deep learning using Wi-Fi and iBeacon is proposed. Signal strength of each AP and iBeacon is collected at each reference point in offline phase, and
is used to train the stacked auto-encoder which is used to extract features from a large number of signal strength samples with noise. These features are used to construct the
fingerprint database. The features of the point to be measured can be obtained by the stacked auto-encoder in online phase. Then these features are matched in fingerprint database.
The position of the point to be measured is estimated by the nearest neighbor algorithm. Experimental results show that the proposed indoor localization algorithm has higher
localization accuracy.

Key words: indoor localization, deep learning, stacked auto-encoder, nearest neighbor algorithm, iBeacon, Wi-Fi