改进LAMBDA算法实现BDS双频整周模糊度快速解算

发布时间：2018-08-07 16:20

【摘要】：利用北斗卫星导航系统进行基线解算时,正确解算模糊度是关键问题.利用北斗卫星导航系统双频载波相位观测值组成宽巷双差观测值,运用目前被认为效果最好的模糊度解算方法——LAMBDA算法解算宽巷双差整周模糊度,但是,该算法存在模糊度浮点解偏差大,搜索范围大的缺点.因此,该文提出利用Tikhonov正则化改进LAMBDA算法,对宽巷双差观测方程中未知参数的系数矩阵进行奇异值分解,用分解后的协方差矩阵替换LAMBDA算法中的协方差矩阵进行宽巷双差整周模糊度的搜索,提高了宽巷双差整周模糊度浮点解的精度和模糊度固定正确率.改进后的LAMBDA算法模糊度浮点解偏差平均值从1.0829周减少到0.0059周,方差值从4.0534周减少到0.0479周,模糊度固定正确率从93.52%提高到100%.
[Abstract]:When using Beidou satellite navigation system to calculate the baseline, the correct ambiguity resolution is the key problem. The dual-frequency carrier phase observations of Beidou satellite navigation system are used to make up the double-difference observation values of wide roadway, and Lambda algorithm is used to calculate the integer ambiguity of double difference in wide lane, which is considered to be the most effective ambiguity resolution method at present. The algorithm has the disadvantages of large ambiguity floating point solution deviation and large search range. Therefore, an improved LAMBDA algorithm based on Tikhonov regularization is proposed to decompose the coefficient matrix of unknown parameters in the two-difference observation equation of wide roadway by singular value decomposition. The covariance matrix in LAMBDA algorithm is replaced by the decomposed covariance matrix to search the integer ambiguity of double difference in wide lane, which improves the precision of floating point solution and the fixed accuracy of ambiguity. The improved LAMBDA algorithm reduces the average ambiguity floating-point deviation from 1.0829 weeks to 0.0059 weeks, the square difference from 4.0534 weeks to 0.0479 weeks, and the accuracy of fixed ambiguity from 93.52% to 100 weeks.
【作者单位】：辽宁工程技术大学测绘与地理科学学院;
【基金】：国家自然科学基金(41474020,41504010)~~
【分类号】：TN967.1

【相似文献】