基于Wi-Fi指纹的改进型室内定位算法研究

Research on improved indoor location algorithm based on Wi-Fi

针对复杂多变的室内环境,导致运用Wi-Fi位置指纹定位算法定位的确定性和抗干扰能力较差,且定位精度及定位效率无法得到保证的问题,提出了一种改进型的位置指纹识别算法,将手持设备的方向角和接入点的数量加入离线指纹数据库,在线定位阶段采用K-means聚类算法与加权KNN算法相结合进行位置估计。通过实验验证得出,改进后的位置指纹定位方法与常规的位置指纹定位算法相比,定位精度得到明显提高,效率显著加快。

The current indoor environment is very complicated and changeable,which leads to poor certainty and anti-jamming ability of Wi-Fi location fingerprinting algorithm,and the accuracy of positioning and the efficiency of positioning can not be guaranteed. In this paper,an improved location fingerprinting algorithm is proposed based on the above problems. The orientation angle and the number of access points of handheld devices are added to the offline fingerprint database. K-means clustering algorithm and weighted KNN algorithm are used in the online positioning phase Location estimation. The experimental results show that compared with the conventional location fingerprinting algorithm,the improved location fingerprinting method can significantly improve the positioning accuracy and speed up the efficiency significantly.