Adaptive Kalman filter based on variance component estimation for the prediction of ionospheric delay in aiding the cycle slip repair of GNSS triple-frequency signals
Adaptive Kalman filter based on variance component estimation for the prediction of ionospheric delay in aiding the cycle slip repair of GNSS triple-frequency signals
复制标题
基于方差分量估计的自适应卡尔曼滤波器预测电离层延迟以帮助 GNSS 三频信号周跳修复
DOI：
10.1007/s00190-018-1116-4
复制
发表时间：
2018
影响因子：
4.4
通讯作者：
Wang Qianxin
中科院分区：
文献类型：
--
作者：
Chang Guobin;Xu Tianhe;Yao Yifei;Wang Qianxin
为了利用电离层延迟的时间平滑性来辅助周期滑动检测，提出了一种基于方差分量估计的自适应卡尔曼滤波器。在有色测量噪声滤波算法的开发中，充分考虑了相邻时代测量值之间的相关性。在这个滤波框架内，预测了电离层的时代性延迟。利用这一预测，对全球卫星导航系统三频信号的潜在周期跳进行了修复。循环卡瓦是逐步修复的；即，首先用于两个额外宽的车道组合，然后用于第三个频率。在第三频率的估计中，采用随机模型，考虑电离层延迟预测误差与差时相位测量误差之间的相关性。所提出方法的实现细节以表格形式列出。利用实际的北斗卫星导航数据集验证了所提方法的性能。大多数周期滑移，无论大小，都能以浮点值的形式估计出令人满意的高精度，因此通过简单的舍入就能正确地得到它们的整数值。更具体地说，所有人工引入的非平凡循环卡瓦都得到了正确的修复。
In order to incorporate the time smoothness of ionospheric delay to aid the cycle slip detection, an adaptive Kalman filter is developed based on variance component estimation. The correlations between measurements at neighboring epochs are fully considered in developing a filtering algorithm for colored measurement noise. Within this filtering framework, epoch-differenced ionospheric delays are predicted. Using this prediction, the potential cycle slips are repaired for triple-frequency signals of global navigation satellite systems. Cycle slips are repaired in a stepwise manner; i.e., for two extra wide lane combinations firstly and then for the third frequency. In the estimation for the third frequency, a stochastic model is followed in which the correlations between the ionospheric delay prediction errors and the errors in the epoch-differenced phase measurements are considered. The implementing details of the proposed method are tabulated. A real BeiDou Navigation Satellite System data set is used to check the performance of the proposed method. Most cycle slips, no matter trivial or nontrivial, can be estimated in float values with satisfactorily high accuracy and their integer values can hence be correctly obtained by simple rounding. To be more specific, all manually introduced nontrivial cycle slips are correctly repaired.