利用现有无线局域网进行室内定位算法研究

目前，多种WiFI室内定位方案被提出，但是往往需要重新部署无线AP，造成成本和复杂度上升。本文充分利用现有无线局域网的拓扑结构进行室内定位研究，提出了一种自适应网络变化的WKNN指纹算法，该算法通过实时监控无线AP的匹配数，自动根据位置适应网络变化，定位精度明显提高。在此基础上，为了减少无线信号不稳定引起的定位误差，提出了一种新的数据修正方法，该方法根据移动平均速度动态预测标准，动态调整a参数将预测坐标与实测坐标加权，从而得到最终定位坐标。最后，算法在实际环境中验证表明，利用现有无线局域网的自适应网络算法和数据修正使定位获得了33.5%的误差改善。 

A Study on Indoor Localization Algorithm Using Existing Wireless Local Area Network

At present, variety of WLAN indoor positioning schemes have been proposed with redeploying extra wireless AP(Access Point), leading to the rise of costs and complexity. This paper make full use of existing wireless local network for indoor positioning research. In this paper, an adaptive network WKNN(Weighed K-nearest Neighbor) algorithm that matching number of APs (Access Points) was monitored real-timely to adapt to the changes of network was proposed, greatly optimizing the fingerprint algorithm. After that, on the basis of the adaptive algorithm, in order to decrease the positioning error caused by the instability of wireless signal, we provided a data correction method. In this method, the final localization coordinate was the average of predicted coordinate and measured coordinate weighted by a certain coefficient . Finally, our proposal has been tested in real environment, and verification of experiment shows that localization accuracy has been improved obviously with 33.5%.