基于EEMD的全质心UWB室内定位算法

针对室内环境下的非视距(non-line-of-sight,NLOS)及多径(Multipath)传播给定位结果带来较大误差的问题,本文先将通过超宽带(ultra wideband,UWB)室内定位系统中的TOA模型测得的目标节点(携带定位终端的人或物体)分别到三个锚节点(下位机)的距离值进行集合经验模式分解(Ensemble Empirical Mode Decomposition,EEMD)分析除噪,再运用三角形全质心定位算法求出目标节点在二维坐标系中的坐标值,最后从这一组定位结果中随机选一个坐标值作为目标节点最终的定位结果。结果表明:与传统的粒子滤波法、贝叶斯滤波法、BP神经网络法、泰勒级数法、小波滤波法等提高定位精度的算法相比,该算法提高了定位精度及鲁棒性等性能。

UWB indoor location algorithm based on EEMD

When determining the object or person＇s position in the indoor environment, a novel Full Centroid Position scheme based on noise removed by EEMD in indoor position system is proposed by researching the NLOS and multipath transmission in indoor environment and the distance test models of ultra wideband（UWB）.In this scheme, the distance between the target nodes and the anchor nodes is measured by the model of TOA（time of arrival）,in order to remove the noise added to the distance values, EEMD analysis filter will be used. And then, the values produced by the filter will be used by Full Centroid Position scheme to calculate the position results, at last a random coordinate from the results will be regarded as the final position result. The results show that when compared with the traditional Particle Filter scheme, Bayes Filter scheme, BP neural network scheme,Taylor series expansion scheme, Wavelet Filter scheme and other schemes to improve the precision, the algorithm has the characteristics of high precision, strong robustness.