联合Spline插值的Wi-Fi指纹匹配定位算法

Wi-Fi Fingerprint Localization Uniting Spline Interpolation

该文从降低现有的Wi-Fi室内定位技术算法成本、保证定位精度的角度出发,提出了联合Spline插值的Wi-Fi指纹匹配定位算法。在构建信号强度指纹数据库方面,该文提出了稀疏指纹库的构建,大大降低了数据采集的工作量和硬件需求,并且提出将混合滤波与Spline插值方法结合,对稀疏指纹数据库进行丰富。在信号强度指纹数据库的插值方面,经过相同程度的混合滤波后,与已知的反距离加权(IDW)插值算法相比,联合Spline插值方法能够实现对数据库的精确填充,实现更高的定位精度;在指纹匹配定位方面,采用K最邻近(KNN)等匹配算法实现高精度定位。通过仿真实验证明,该文所提出的联合Spline插值的Wi-Fi的指纹定位方法能够在仅需要构建低成本稀疏指纹数据库的前提下,保证较高的定位精度。

In order to reduce the cost of the existing Wi-Fi indoor positioning technology algorithm and ensure the positioning accuracy,a Wi-Fi fingerprint matching positioning algorithm uniting Spline interpolation is proposed in this paper.In terms of constructing the signal strength fingerprint database,the construction of a sparse fingerprint database is proposed,which greatly reduces the workload and hardware requirements of data collection.In addition,the combination of hybrid filtering and spline interpolation method is proposed to enrich the sparse fingerprint database.In terms of interpolation of the signal strength fingerprint database,after the same degree of hybrid filtering,compared with the known Inverse Distance Weighting(IDW)interpolation algorithm,the spline interpolation method can accurately fill the database and achieve higher positioning accuracy.In terms of fingerprint matching and positioning,matching algorithms such as K-Nearest Neighbor(KNN)are used to achieve high-precision positioning.Simulation experiments show that the proposed Wi-Fi fingerprint positioning method uniting spline interpolation can ensure high positioning accuracy under the premise of only building a low-cost sparse fingerprint database.