基于不等式约束算法改善高纬度地区GNSS电离层建模精度

Inequality Constraints Algorithm to Improve the High Latitudes GNSS Ionosphere Modeling Accuracy

由于当前国际GNSS服务组织(IGS)跟踪站在高纬度区域分布不均,用于电离层拟合建模的观测数据不均匀、不全面,导致在利用这些IGS跟踪站观测数据建立电离层拟合模型时该区域的电离层电子总含量拟合模型精度不够,电离层拟合模型格网输出值甚至出现了大量有悖于电离层实际物理意义的负值和零值现象。针对此问题,利用不等式约束算法,对拟合模型电离层格网输出的负值点、零值点加入不等式约束条件,采用附加不等式约束条件的最小二乘法对参数的解算进行优化,并用实测GNSS数据进行验证。实验结果表明,该算法对测站分布稀疏的高纬度地区出现大量零值和负值的情况有明显改善。此外,该算法对全球电离层模型的建模精度也有一定程度的提高。

Due to uneven distribution of the international GNSS service(IGS)organization tracking stations in high latitudes,the observation data used for ionospheric fitting modeling are uneven and incomplete,leading to inadequate modeling accuracy of the total ionospheric content in this region.When the ionospheric fit model is established using observation data from these IGS tracking stations,the grid output values of the ionospheric fitting model contain a large number of negative and zero values,which are contrary to the actual physical meaning of the ionosphere.Aiming at this problem,we adopt the ionospheric grid output negative point,zero point to join inequality constraint conditions,and the inequality constraint least square method,to calculate parameters of optimization.Experimental results show that the algorithm shows a significant improvement in reducing the large number of zeros and negative values in the sparse high-latitude areas,and modeling accuracy of the global ionosphere model is also improved.