基于经验模态分解的周跳探测

Cycle-Slip Detection Based on Empirical Mode Decomposition

提出了一种基于经验模态分解(EMD)对载波相位测量进行周跳探测的新方法。利用非差和双差组成检验量,对采样率分别为1s和30s的观测值进行周跳探测,并与小波分析法进行对比。实验表明,经验模态分解法对于10周以内的周跳探测可以取得满意的效果。相对于小波分析法探测周跳,经验模态分解可以做到自适应,解决了由于小波基的选取不当对周跳探测结果的影响。并且进一步分析了EMD的端点效应对周跳探测的影响。

A new method for detecting cycle slips of the carrier phase measurements based on empirical mode decomposition （EMD） was proposed. The non-differential and double differential test volume were constructed by using carrier phase measurements of 1 s and 30 s sampling interval, and were detected based on EMD and wavelet. The experiments showed that the good result using EMD method was obtain if the cycle slip was within 10. Relative to the wavelet method, EMD did not need to select specific parameters, which avoided the shortcoming caused by selecting not appropriate wavelet base. Further more, endpoint effects of EMD method was analyzed and some draw backs of the EMD method in detecting cycle slip were also pointed out.