高维情况下双差整周模糊度LAMBDA法解算分析

Analysis of a double-differ ambiguity solution by the LAMBDA method in high dimensional application

为了研究在多基线和多定位系统网络式等高维应用情况下整周模糊度解算LAMBDA(least-squares ambiguity decorrelation adjustment)方法的失效问题,针对算法的去相关处理过程深入分析,在阐述去相关具体实现流程的基础上,将解算步骤细化.运用矩阵分析的方法,研究了去相关过程中矩阵分解和排序问题,指出了高维情况下模糊度的条件方差在去相关处理后降序排列不完全的问题.从理论角度解释条件方差降序排列是模糊度正确解算的前提,进而得出结论:高维情况下整周模糊度解算失效的原因是条件方差降序处理不完善.

Invalidation problems of the least-squares ambiguity decorrelation adjustment（LAMBDA） method in high dimensional applications such as the model of multi-baselines and networks were researched.Based on the expatiation of the process of LAMBDA decorrelation,a problem relating to decomposition and ordering of the matrix was analyzed using the matrix analysis method.Also,the incomplete ordering of conditional variances after the decorelation of high dimensional ambiguity was noted.Furthermore,the descending order of conditional variances was theoretically shown to be the premise of ambiguity valid solutions.The conclusion was drawn that the reason for the disabled LAMBDA decorrelation was that the descending management of conditional variances was faulty.