单频载波相位的周跳探测与修复算法研究

Algorithm of cycle-slip detection and correction in single-frequency carrier phase

为了获得高精度的导航定位结果,必须对载波相位中的周跳进行快速准确地探测与修复。分析了周跳产生的原因及特点,将周跳看作载波相位测量数据的奇异点,在此基础上提出了一种新的载波相位周跳探测与修复方法。首先对载波相位测量数据进行4次差分预处理;然后对差分序列进行小波变换,通过小波系数的模量极大值点的位置准确探测出周跳发生的历元;最后对周跳发生前的相位数据建立基于经验模式分解与径向基神经网络的组合预测模型,通过对比预测值与实际测量值的大小来修复周跳。应用实际数据对算法进行验证,结果表明:新方法适用于单频载波相位数据,可以对1周以上的周跳进行准确探测与修复。

In order to obtain high precision positioning and navigation results with GPS,cycle-slip in carrier phase observations must be correctly and quickly detected and corrected.Based on the characteristics of cycle-slip,the location of the cycle-slip is regarded as a singular point of the signal,and then a new method for detecting and correcting cycle-slip is proposed.Firstly,the fourth difference series of carrier phase observations are calculated.Secondly,wavelet transform is carried out on the difference series,and the location of cycle-slip can be detected by ascertaining the point of modulus maximal value of the wavelet coefficients.Finally,a combined forecast model of carrier phase observations based on empirical model decomposition and RBF neural network is established.The number of cycle-slip can be determined by comparing the forecasted and actual values.Experiment results show that the method can be used in single-frequency observations and improve the detection and correction efficiency of cycle-slip.The minimum value of cycle-slip that can be identified is 1 cycle.