【摘要】 本文在综合分析了各种周跳探测算法的基础上,针对参考站网络周跳探测的特点开发出一种参考站网络实时探测周跳的新算法。该算法使用双差宽巷和(9,-7)消电离层线性组合、参考站精确坐标和精密星历来构造观测量,卡尔曼滤波来估计电离层参数,并构造基于χ2分布的整体和局部统计量在给定的显著水平下来同时探测多颗卫星的周跳;采用产生周跳候选值、筛选和验证周跳候选值的方法对周跳进行修复。最后,以四川综合GPS服务网(SIGN)数据对该算法进行了验证。实验结果表明,该算法在100多个历元的长数据缺口以及1周小周跳发生的情况下同样具有很好的效果,且具有很高的效率和可靠性。更多还原

【Abstract】 Based on a comprehensive study of various algorithms presented in most available literatures,a new algorithm for cycle slip on-line detection has been developed regarding to the characteristics of reference station network.This proposed algorithm uses double-difference wide-lane and(9,-7) linear combinations,as well as the precise coordinates of reference stations and precise ephemeris to construct observation equation.And it also designs Kalman Filter to estimate the ionosphere parameters.Subsequently,a global and local statistic test is carried out to detect cycle slips over multi satellites simultaneously based on χ~2 distribution at a given confidence level.Besides,a procedure of cycle-slip candidate generation,filtering and validation is performed to repair the cycle slip.Finally,a validation procedure is implemented to validate the proposed algorithm using the data of Sichuan Integrated GPS Network(SIGN).The experimental results show the efficiency and reliability of the mentioned algorithm,even under the circumstances of a long data gap as long as about 100 epochs and cycle slips as small as one cycle.更多还原