一种基于传统MART的电离层层析新算法

A New Algorithm of Ionospheric Chromatography Based on Traditional MART

层析算法研究是电离层层析技术的关键。针对传统的乘法代数重构算法(MART)存在精度不高的问题,提出了一个改进的电离层层析算法(IMART)。该算法一方面将像素内的截距值与电子含量作为组合量,构造了新的组合量;另一方面改变了松弛因子的获取方式,削弱了噪声对反演结果的影响。利用GPS实测数据,对该算法的可靠性和优越性进行了仔细验证。根据统计分析和定量比较的结果,发现该算法表现出显著的重构精度且相较于其他经典层析算法有18%的提高。

The study of tomographic algorithms is the key to ionospheric tomographic. An improved ionospheric tomography algorithm (IMART) is proposed, which is based on the low accuracy of the traditional multiplier algebraic reconstruction algorithm (MART). The intercept value and the electronic content in the pixel are used as the combined quantity to construct a new combination quantity in the algorithm. On the other hand, the acquisition method of the relaxation factor is changed and the influence of noise on the inversion result is weakened. Then using GPS measured data, the reliability and superiority of the algorithm are carefully verified. According to the results of statistical analysis and quantitative comparison, the algorithm shows significant reconstruction accuracy and has 18% improvement compared with other classical tomographic algorithms.