EMD-RLS联合滤波算法及其在北斗多路径削弱误差中的应用

EMD-RLS combination algorithm and its application in weakening BDS multipath error

多路径误差是全球导航卫星系统高精度定位的主要误差源。针对北斗导航定位系统的多路径特性,结合经验模态分解(EMD)和递归最小二乘算法(RLS)算法的优势,提出了一种EMD-RLS联合滤波算法来削弱北斗多路径特性的影响。首先,利用EMD将原始信号分解成固有模态函数(IMF),利用两个指标将其细化成三类,即噪声IMFs、混合IMFs和信息IMFs;然后,利用RLS算法对混合IMFs进行滤波,将经RLS滤波后得到的“干净”的数据与信息IMFs进行重构,最终达到去噪结果,即得到北斗多路径误差改正模型。仿真数据和北斗实测数据的处理结果表明:EMD-RLS算法相对于EMD和RLS算法,降噪效果更好;其建立的多路径误差改正模型有效地削弱多路径效应的影响,N、E、U三个方向精度分别提高了39%、70%、58%。

Multipath error is the main error source for high-precision positioning of global navigation satellite system.Aiming at the multipath characteristics of Beidou navigation and positioning system,and com- bined with the advantages of the empirical mode decomposition (EMD) and the recursive least-squares algorithm (RLS),An EMD+RLS combined filtering algorithm is proposed to weaken the influence of BeiDou multipath characteristics.EMD is applied to decompose the coordinate sequence into an intrinsic mode function (IMF),which is then divided into three categories (i.e.noise IMFs,mixed IMFs,and information IMFs) by using two indices.The RLS algorithm is used to process the mixed IMFs.The “clean” data and information IMFs obtained after RSL filtering are reconstructed to obtain the BeiDou multi-path error correction model. The processing results of simulation data and BDS measured data show that the EMD-RLS algorithm has better noise reduction effect than the EMD and RLS algorithms,and the improvement of BDS positioning result is more significant.The established multipath error correction model effectively weakens the influence of multipath,and the accuracy of N,E,and U directions is increased by 39%,70%,and 58%,respectively.