摘要
针对室内定位中,受无线通信距离和环境的影响,紫蜂(ZigBee)观测信号的随机噪声较为显著,导致定位结果的精度和稳定性较弱的问题,提出一种融合迭代平均滤波的RSSI室内定位算法:在数据采集过程中,采用高斯滤波对观测信号进行预处理,并根据接收信号强度指数测距理论建立起路径损耗模型,由此估算出信号接收节点到广播信标节点间的距离;然后,在定位解算过程中,在线性最小二乘算法的基础上融合迭代平均滤波算法,以抑制随机误差对定位结果的影响;最后,采用自研的CC2530 ZigBee模组对算法的可行性和稳定性进行实验验证。结果表明,等权最小二乘算法的点位精度约为0.64m,距离定权后,最小二乘算法的精度提升到0.52m;用等权最小二乘算法融合迭代平均滤波,精度也能达到0.5 m左右,与距离定权最小二乘的精度相当;同时,融合距离定权和迭代平均滤波2种方法的点位精度可以提高到0.35 m左右。
Aiming at the problem that during indoor positioning,affected by wireless communication distance and environment,ZigBee observation signal has significant random noise,which leads to certain deficiencies in the accuracy and stability of positioning results,the paper proposed a RSSI indoor location algorithm based on iterative mean filter:Gaussian filter was used to eliminate the influence of low probability observation events in the process of data acquisition,and the path loss model was established according to received signal strength indicator ranging theory,so as to estimate the distance between signal receiving node and broadcast beacon node;then,in the process of positioning solution,the iterative mean filter algorithm was integrated on the basis of the linear least squares algorithm to suppress the influence of random errors on the positioning results;finally,the self-developed CC2530 ZigBee module was used to verify the feasibility and stability of the algorithm.Results showed that the point position accuracy of the equal-weighted least squares algorithm would be about 0.64 m,and the accuracy of the least squares algorithm could be improved to 0.52 m after distance-weighted determination;by integrating the iterative mean filter into the equal-weighted least squares algorithm,the accuracy could reach about 0.5 m,which would be equivalent to that of the distance-weighted least squares;at the same time,the point accuracy of integrating two methods of the distance-weighed and the iterative mean filter could be improved to about 0.35 m.
作者
潘兴驰
闫文林
PAN Xingchi;YAN Wenlin(School of Geography,Geomatics and Planning,Jiangsu Normal University,Xuzhou,Jiangsu 221116,China)
出处
《导航定位学报》
CSCD
北大核心
2024年第2期79-85,93,共8页
Journal of Navigation and Positioning
