EEMD及LS-SVM在单频周跳探测与修复中的应用

Study on cycle-slip detection and repair in single-frequency based on EEMD and LS-SVM

针对北斗单频观测数据中微小周跳难以探测与修复的问题,采用基于总体经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)及最小二乘支持向量机(Least Squares Support Vector Machine,LS-SVM)的方法用于周跳探测与修复。该方法首先利用伪距减相位历元间差分构造周跳检测量,将其进行EEMD分解获得若干本征模式分量(Intrinsic Mode Function,IMF);然后利用相关性分析选取周跳信息明显的IMF分量信号并对其进行Hilbert谱分析,根据Hilbert谱中模极大值点的位置探测出周跳发生的历元;最后对筛选出的IMF分量信号利用LS-SVM预测,通过比较实测值与预测值的大小来修复周跳。采用实测数据验证本文方法,实验结果表明:本文方法可以有效探测并修复北斗单频观测数据中出现的1周左右的微小周跳。

Aiming at the problem that the small cycle-slip of BDS single-frequency observation data is hard to detect and repair simultaneously, a method based on Ensemble Empirical Mode Decomposition（EEMD） and Least Squares Support Vector Machine （LS-SVM） is used in this paper. Firstly, this method constructs the cycle-slip detectable quantity by difference of the pseudorange and the carrier-phase between adjacent epoch, decomposes it by EEMD to get a certain number of Intrinsic Mode Function（IMF） components; Secondly, using cross correlation analysis to screen out the IMF components, detecting the epoch where cycle-slip appears by calculating the maximum point of Hilbert amplitude spectrum of the IMF components witch are selected; Finally, in order to predict the cycle-slip, the LS-SVM is employed to model the IMF components including the cycle-slip information before the epoch where cycle-slip not occurs, the cycle-slip is repaired by comparing the predicted value with the real one. An experiment was carried out to verify the proposed method using the real tested data, the results indicate that the method in this paper can effectively detect and repair 1 cycle-slip.