【作者】 徐霖；

【导师】 郭立新；

【作者基本信息】 西安电子科技大学 ， 无线电物理， 2008， 硕士

【摘要】 本文首先应用Monte Carlo方法(线性滤波法)模拟生成海面表面轮廓,利用迭代加速算法MOMI方法计算了海面的双站散射截面,并且分析了风速和入射角等参数对海面双站散射截面的影响。基于RBF人工神经网络实现了海面散射场的建模。利用基于电流计算的矩量法结合高频算法基尔霍夫近似的混合算法分析了高斯随机粗糙面以及一维PM谱粗糙海面及其上方二维无限长任意截面导体目标的双站复合电磁散射特性,并且利用结合互易性定理和MOMI方法的混合方法讨论了粗糙海面与其上方三维平板目标的复合电磁散射问题。训练RBF神经网络进行复合电磁散射建模并对一定风速、入射角、频率和目标高度、目标半径等参数下的复合散射场进行推测,预测值同目标数据之间误差较小,结果令人满意。更多还原

【Abstract】 In this paper, the generally statistical method - Monte Carlo Method is used togenerate the sea rough surface, the bistatic RCS for the sea surface is calculated usingthe MOMI (Method of Ordered Multiple Interactions), and the influence of theparameters such as the speed of wind and the incidence angle to the bistatic RCS areanalyzed. Modeling for the scattering field of the sea surface is implemented with theRBF neural network. The composite scattering from two-dimensional infinitely longconducting target with arbitrary cross section above Gaussian rough surface andone-dimensional sea surface with PM spectrum are calculated using the hybrid methodcombining MOM with Kirchhoff Approximation (KA). The electromagnetic scatteringfrom three-dimensional plate target above sea surface is investigated by employing thehybrid method combining the reciprocity theorem and MOMI method. The RBF neuralnetwork is trained for the modeling of composite electromagnetic scattering and thecomposite scattering field under different sea/target parameters such as the wind speed,the incidence angle, the incident frequency and the height and size of the target isestimated. The error between the network consequence and the target data is in asatisfied range.更多还原