基于实测数据的WiFi室内定位技术研究

Research on WiFi Indoor Positioning Technology Based on Measured Data

室内定位技术的发展给生活带来了便利。WiFi位置指纹室内定位技术在室内定位技术领域较为普遍,但是原始的数据经过简单的滤波后使用传统的WKNN算法进行匹配定位,得到的最终定位点位置存在一定误差。文章通过在数据预处理阶段对Kmeans算法进行优化,其均方根误差较传统Kmeans算法的均方根误差降低了3.2%,匹配定位阶段匹配算法对KWNN算法的权值进行方差优化,其平均定位误差较传统KWNN算法的平均定位误差降低了23.6%。

The development of indoor positioning technology brings convenience to life.WiFi location fingerprint indoor positioning technology is more common in the field of indoor positioning technology,but the final positioning point position got from the original data,has a certain error after simple filtering with the traditional WKNN algorithm for matching positioning.Kmeans algorithm is optimized in the data preprocessing stage.Its root mean square error is decreased by 3.2%compared with that of the traditional one.The matching algorithm in matching positioning stage performs variance optimization on the weight of KWNN algorithm.The average positioning error is 23.6%lower than that of the traditional one.