狄克逊检验法滤波RSSI的室内定位算法  

Indoor Positioning Algorithm Based on Dixon Test Filtering RSSI

作　　者：王建强[1] 代阳 雷倩芳 WANG Jianqiang;DAI Yang;LEI Qianfang(Faculty of Geomatics,East China University of Technology,Nanchang Jiangxi 330013,China)

机构地区：[1]东华理工大学测绘工程学院,江西南昌330013

出　　处：《传感技术学报》2021年第1期118-123,共6页Chinese Journal of Sensors and Actuators

基　　金：江西省自然科学基金项目(20202BABL202046);国家自然科学基金项目(41464001);国家重点研发计划项目(2016YFB0501405)。

摘　　要：针对接收信号强度指示(Received Signal Strength Indication,RSSI)测距定位算法存在定位结果不稳定且精度低的问题,本文分析了一种基于狄克逊检验法滤波RSSI的高斯牛顿定位(Dixon test filter RSSI Gauss-Newton,DF-RSSI-GN)算法。采用狄克逊(Dixon)检验法滤波剔除观测信号异常值使得观测数据偏度降低,根据偏态程度对观测信号进行高斯均值滤波并通过非线性回归模型拟合RSSI衰减模型参数,在目标点坐标求解阶段利用滤波后的观测信号确定不同方向上的权值进行高斯牛顿(Gauss-Newton)迭代定位。实验结果表明,DF-RSSI-GN算法定位平均精度在1.5 m左右,相比RSSI定位算法和最小二乘定位算法,精度提高1倍以上。In order to solve the problem of unstable location results and low accuracy of Received Signal Strength Indication(RSSI)algorithm,this paper analyzes a Gauss-Newton localization algorithm based on the Dixon test filtering RSSI(DF-RSSI-GN).The method of Dixon test is used to filter out the observation signal outliers,which can reduce the deviation of the observation data.According to the degree of deviation,the observation signal is filtered by Gaussian-mean and the parameters of the RSSI attenuation model are fitted by nonlinear regression.The weights in different directions are determined by the filtered observation signal to carry out Gauss-Newton iterative positioning.The experimental results show that the average positioning accuracy of DF-RSSI-GN algorithm is about 1.5 m,which is more than one time higher than these positioning algorithms of RSSI and Least Squares.

关 键 词：室内定位 狄克逊检验法滤波 高斯牛顿法 RSSI 

分 类 号：P228[天文地球—大地测量学与测量工程]