分析了传统单频数据预处理方法在复杂运动状态下的缺陷,打破常规周跳探测基于时间序列的处理思维,提出基于双差的单频单站数据预处理方法。主要内容包括基于双差验后单位权中误差检测周跳发生的时刻;基于稳健估计的残差检测周跳发生的卫星;利用周跳回代估计进行周跳修复。通过模拟特定运动状态的动态数据及实测地震数据,验证了该方法在接收机复杂运动状态中的有效性及连续周跳探测的能力。
This paper analyzes the drawbacks of traditional single-frequency preprocessing method, only by analyzing time series, in complex movement and proposes single-frequency preprocessing method based on double differences. The method includes the following aspects, detect epochs of cycle slip based on posteriori standard error of unit weight of double difference, recognize satellites of cycle slip based on residuals of robust estimation and fix cycle slip by back-substitution of cycle slip. Finally, this paper investigates the efficiency of the new method in compound movement of receivers and its capacities to detect continuous cycle slips by simulating kinematic data and analyzing real earthquake data.