一种基于Voronoi图和朴素贝叶斯的室内定位无线地图构建方法

A Novel Radio Map Construction Method for Indoor Localization Based on Voronoi Diagram and Bayesian Approach

室内无线定位以其低成本、高普适性等优点成为用户定位研究领域的热点,针对现有基于位置指纹数据库的室内无线定位算法,因室内环境复杂存在定位精度低、数据通信能耗大等问题,提出一种Voronoi图和朴素贝叶斯定位相结合进行无线地图(指纹数据库)构造的算法模型。首先,建立两级校准点模型,一级校准点为预先选择且进行收集RSSI(Received signal strength indicator)样本,在此基础上通过无线传播模型计算得到二级校准点;然后,对无线地图进行Voronoi区域生成,每个Voronoi图包含一个一级校准点和多个二级校准点;最后,在匹配得到的Voronoi区域内运行朴素贝叶斯算法,获得定位目标的估计位置。实验结果表明,提出的算法模型能降低收集RSSI数据的成本,同时提高目标定位精度,具有一定的应用价值。

Indoor localization using the fingerprinting technique, namely, the radio map, has attracted much attention in the research community recently. However, constructing a complete radio map is extremely labor-extensive and time-consuming, especially for a wide area. Although some works have been done to reduce the number of calibration points, the accuracy decreases if there are not enough fingerprints. In this paper, we propose a novel method based on the radio propagation model to construct a radio map with full fingerprints. In the radio map, the calibration points （CPs）, i. e. , fingerprints, are classified into two categories： the primary CPs chosen to collect received signal strength Indicator （RSSI） artificially, and the secondary CPs obtained through some calcula tion. Based on the radio map constructed, we employ the Voronoi diagram to divide the area into several Voronoi regions and restrict the localization algorithm to run in a specific Voronoi region to reduce computational complexity. The comparison results show that our method saves a lot of time and human effort in collecting RSSI samples, and achieves much higher accuracy than other existing schemes.