以各向异性材料的散射问题的正演算法(FDTD)作为基础,利用机器学习方法——支持向量机对各向异性材料电磁参数进行了反演研究。以各向异性介质球的雷达散射截面(RCS)作为训练样本信息,经过支持向量机适当的训练,实时重构了各向异性介质球的相对介电常数和电导率。与此同时,还给出了利用BP神经网络方法反演的结果,显示了该方法的有效性和准确性。 Based on the forward algorithm (FDTD) of scattering of anisotropic materials, the electromagnetic parameters of anisotropic materials are inversely studied by means of machine learning method and support vector machine. Using the RCS of anisotropic medium sphere as the training sample information, the relative permittivity and conductivity of the anisotropic medium sphere are reconstructed in real time through appropriate training of SVM. At the same time, the results of BP neural network inversion are also given, which shows the effectiveness and accuracy of the method.