摘要
针对室内信号时变性导致定位不准的问题,提出了一种改进的3阶段位置指纹定位法。采样阶段,将采集信号的坐标、方位、接收信号强度的高斯分布及其对应的无线接入点等信息存储在数据库中生成位置指纹;在校正阶段中,利用参考点间信号强度的关联性信息,使用局部加权线性回归法,计算出一些虚拟点的信号强度;最后是线上实时定位阶段。通过与传统的加权K最邻近算法、直方图和联合聚类等3种定位方法相比较,该算法在同样的场景下可以取得更好的定位精度。
A 3-step location fingerprint （LF） method for the inaccuracy problem caused by the time- varying characteristic of signal in indoor wireless local area networks （WLAN） is proposed. Sampling step, during which a radio map is built： storing calibration points （CPs） in a database, defined by their coordinates, orientation and set of RSS Gauss distributions linked to the source access points （APs） ; in the second step, an augmented radio map is created using locally weighted linear regression by leveraging RSS relations from nearby CP neighbors; the last step is the positioning process performed in real-time. Through the experimentation, the proposed approach is proved to provide better accuracy than weighted KNN method, histogram method and joint clustering （JC） method.
出处
《信息技术》
2013年第7期40-43,共4页
Information Technology
基金
国家自然科学基金资助项目(50505039)
