一种附加尺度因子的电离层层析新方法

用乘法代数重构方法反演电离层电子密度时，没有GPS测量射线穿过的电离层网格的反演精度依赖于迭代初值。就此，提出一种改进的代数重构算法--附加尺度因子的代数重构算法（Scaling-ART）。将实测STEC与迭代中当前电子密度计算的STEC的比值作为基本尺度，依据电子密度的空间相关性和网格中心到观测射线的距离，对没有信号穿过的网格也计算出对应尺度进行调整，使其向真值逐渐逼近，提高反演精度。实验证实，改进的新算法相比于原有的乘法代数重构算法能够取得更好的反演精度。

A New Algorithm for Ionosphere Tomography by Using Scaling Factor

Using the multiplicative algebraic reconstruction technique(MART), the accuracy of ionospheric electron density in the cells without any rays traversing them depends on the initial value, which is generated by empirical model. To solve the above problem, an improved algebraic reconstruction technique is presented in the paper. In this algorithm, a basic scale is calculated as the proportion of observed STEC and reconstructed STEC in some iteration. According to the spatial correlation of electron density, scaling factors are calculated as a function of basic scale and the distance from the center of cell and GPS ray for each cell in the reconstruction area. So, the ionospheric electron density of each cell will get closer to the ground truth. Numerical experimental and actual observation tests prove that the new algorithm is superior to the original MART.