基于BP神经网络的介质圆柱体逆散射方法研究

Inverse scattering by dielectric circular cylinder based on BP neural networks

将人工智能技术应用于介质圆柱体电磁逆散射问题研究,通过BP神经网络将原逆散射问题转化为一个回归估计问题,重构了目标的几何与电磁参数。在TM波的照射下,设置多个目标散射场的观测点,以散射场的幅值作为BP网络的输入,相应的几何与电磁参数作为输出,经过适当的训练,建立了介质圆柱体逆散射模型,并以此模型重构了已知探测范围内的介质圆柱体的半径、相对介电常数及电导率。比较结果显示了该方法的有效性和准确性,为目标的实时逆散射研究提供了一种有效方法。

In this paper,an artificial intelligence technology is applied to the inverse scattering problem of dielectric circular cylinder,which is recast into a regression estimation by means of BP neural network,and the geometry and the electromagnetic parameter of objects are reconstructed.The target scattered electric field is measured at some points,and the electric field values scattered by the target are fed into the BP network,in which output are the geometry and the electromagnetic parameter,and after proper training,the inverse scattering model of dielectric circular cylinder is setup and the radius,relative dielectric permittivity and the electric conductivity of target are reconstructed.The results show that the method is effective and efficient to provide highly efficient solution for the realtime inverse scattering of target.