论文部分内容阅读
针对目前利用WiFi信号进行室内定位实时精度较低的问题,该文提出了一种改进的K最近邻算法。由于室内人体走动对于WiFi信号的不规律干扰,使得室内实时定位的精度带有很大的不确定性。为了实时地消除外界干扰带来的误差,改进的K最近邻算法增加了外部节点来监测周围WiFi信号的强度变化,通过将获取的信号强度与指纹数据库中对应节点的信号强度比对,获取差值,并应用于节点周围的客户端,来实时地校正客户端的定位结果。利用此算法在Android平台上的实验表明,该算法定位简单,可以较为明显地改善节点周围2.4m范围内的实时定位精度,使平均精度能提高0.8~1m左右。
Aiming at the problem of low real-time accuracy of indoor positioning using WiFi signals, this paper proposes an improved K nearest neighbor algorithm. Due to the erratic interference of indoor human walking with WiFi signals, the accuracy of real-time indoor positioning is greatly uncertain. In order to eliminate the error caused by outside interference in real time, the improved K nearest neighbor algorithm adds an external node to monitor the intensity change of surrounding WiFi signals. By comparing the obtained signal intensity with the signal strength of the corresponding node in the fingerprint database, Value, and applied to the client around the node to correct the client’s positioning results in real time. Experiments using this algorithm on the Android platform show that the proposed algorithm has a simple location and can obviously improve the real-time positioning accuracy within 2.4m around the node so that the average precision can be increased by about 0.8 to 1m.