基于WiFi与蓝牙的室内定位技术探究

The Study of Indoor Positioning Technology Based on WiFi and Bluetooth

作为GPS定位技术的"最后一公里"问题,室内定位近年来一直是定位领域的研究热点。目前室内定位技术发展较为迅速,各种定位方法层出不穷,但是不同的定位方法必然都存在缺陷。因此,为了取长补短,发挥各方法的优点,提出基于WiFi与蓝牙的融合定位技术。该技术首先使用基于KL散度核函数的位置指纹定位算法,分别测算出WiFi定位技术和蓝牙定位技术中目标点的坐标集合;其次,利用K-means聚类算法对由蓝牙定位技术得到的结果进行聚类,并选择包含点元素最多的作为蓝牙定位结果;最后,将WiFi定位结果与蓝牙定位结果相融合,该方式得出的结果比任何一种单一定位方式都更加精确。

As the last kilometer problem of GPS positioning technology, indoor positioning has been the focus of localization filed in recent years. At present, indoor positioning technology is developing rapidly and various positioning methods emerge in end- lessly. However, different positioning methods have their shortcomings. Therefore, in order to make full use of the advantages of each method, this paper proposes a fusion positioning technology based on WiFi and Bluetooth. The technology first, em- ploys KL divergence kernel function location fingerprinting algorithm to calculate the position results collection of WiFi and Bluetooth positioning technology. Second, we use K-means clustering algorithm to cluster results obtained by Bluetooth positio- ning technology and choose the one contains the most results as the final result. Finally, the WiFi localization results are fused with the Bluetooth localization results, which are more accurate than any single localization method.