通过研究影响代数重构方法的因素,如松弛因子、投影次序、停止规则、非负约束,针对水汽层析方程组的特点,提出一种基于分组排序的水汽层析约束ART算法(WVTART),并给出投影排序的去相关原则。实验分析证明,WVTART算法提高了收敛速度以及层析结果的精度。
Through the study of relaxation parameter, access order, stop rule, non-negativity constraints, as well as the characteristics of water vapor tomography, this paper proposes water vapor tomography constraint ART algorithm (WVTART) based on grouping and sorting. In order to obtain the global solution efficiently and robustly, observation equations should be sorted by the principle of un-correlation. The results verify that the WVTART method can improve the convergence speed and the accuracy of the solution.