摘要
本文以时域有限差分(FDTD)法数值模拟了各向异性材料的散射特性,在此基础上利用人工智能技术———BP神经网络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.反演结果显示了该方法的有效性和准确性.
This paper is intended to give a presentation of the result of BP neural network method applying to the reconstruction of anisotropic material electromagnetic parameters, which based on finite difference in time domain(FDTD) as the positive algorithm of inverse scattering problems. The BP neural network is trained using different sample data extracted from the radar scattering cross-section of anisotropic material sphere, after a proper training, the aim is to reconstruct the dielectric permittivity and electric conductivity of the unknown anisotropic material sphere. The numerical results show the accuracy and efficiency of the BP neural network method.
出处
《三峡大学学报(自然科学版)》
CAS
2011年第6期110-112,共3页
Journal of China Three Gorges University:Natural Sciences
基金
湖北省教育厅重点资助项目(20111201)
关键词
各向异性材料
电磁参数反演
BP神经网络
雷达散射截面
anisotropic material
reconstruction of electromagnetic parameters
BP neural network
radar cross-section