采用确定性信号传播模型的普适寻优定位方法

Universal optimal localization algorithm by certain signal propagation model

Wi-Fi定位由于不需要额外的基础设施和专门的硬件设备,在室内定位领域有重要应用,不可预测的环境变化引起的接收信号强度(received signal strength,RSS)波动是导致Wi-Fi定位系统精度低的主要原因。提出一种基于测距的普适室内定位方法,使用快速数据聚类训练原始RSS数据,以确保RSS数据的稳定性和有效性;在此基础上,针对提取的RSS信号进行拟合,建立一种确定性信号传播模型,利用天牛须优化方法实现位置求解的高效寻优。通过仿真结果分析,提出的CB-DSPM(clustering by fast search and find of density peaks beetle antennae search-deterministic signal propagation model)定位算法的误差在1.5 m左右,且迭代10~30次之后基本可以收敛到最优位置。

Wi-Fi localization,which is known to be free of extra infrastructure and specialized hardware,has important applications in indoor positioning.However,the received signal strength(RSS)fluctuation caused by unpredictable environmental dynamics is the main reason for the low accuracy of Wi-Fi positioning system.In this paper,a general indoor localization method based on ranging is proposed.Firstly,the original RSS data is trained by fast data clustering to ensure the stability and validity of RSS data.Secondly,a deterministic signal propagation model is established by fitting the extracted RSS signal.Finally,the optimization method of beetle antennae search is used to achieve efficient location optimization.The simulation results show that the error of the proposed clustering by fast search and find of density peaks beetle antennae search-deterministic signal propagation model(CB-DSPM)location algorithm is about 1.5 m,and it can converge to the optimal position after 10-30 iterations.