小波分析和经验模态分解对BDS多路径误差削弱对比研究

Comparative study on wavelet analysis and empirical mode decomposition for BDS multipath error reduction

  • 摘要: 在利用北斗卫星导航系统(BDS)进行高精度变形监测时,BDS信号产生的多路径效应是影响变形监测数据精度和可靠性的一个不可忽视的误差源. BDS有三种不同的轨道卫星,所形成的多路径误差较为复杂. 基于坐标域的多路径误差使用小波分析(Wavelet)和经验模态分解(EMD)进行原始序列降噪,对降噪后序列使用改进恒星日滤波(ASF)进行多路径误差剔除,两种方法分别对基线精度的E方向改善了38.6%和40.8%,N方向改善了59.1%和61.0%,U方向改善了57.8%和57.9%,EMD对坐标序列的平滑和基线精度改善较优.

     

    Abstract: When using BDS for high-precision deformation monitoring, the multipath effect generated by BDS signal is a non-negligible error source that affects the accuracy and reliability of deformation monitoring data. The BDS has three different orbiting satellites, and the resulting multipath error is more complicated. Multipath error based on coordinate domain uses wavelet analysis (Wavelet) and empirical mode decomposition (EMD) for original sequence denoising, and advanced sidereal filter (ASF) for denoising sequence for multipath error culling. The two methods improved the baseline accuracy by 38.6% and 40.8% in E direction, 59.1% and 61.0% in N direction, 57.8% and 57.9% in U direction, respectively. The empirical mode decomposition improved the smoothness of the coordinate sequence and the baseline accuracy better.

     

/

返回文章
返回