欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出一种有效的RBF神经网络二维DoA降维训练方法。利用空间锥角分别对L阵列的两条直线阵进行RBF神经网络模型训练，通过已构建的两个模型对未知来波的空间锥角进行估计，两个空间锥角对应的两个空间半锥面形成的相交线就是来波入射路径。仿真实验结果表明所提方法能有效缩减训练样本集，并能极大降低模型构建的复杂度，而且具备很高的二维来波估计精度，具有广阔的工程应用前景。
英文摘要：
An effective 2D-DoA dimension-degraded training approach based on radial basis function neural networks （ RBFNN ） is proposed in this paper. Two space-conical-angle RBFNN estimation models of two line-arrays of L-shape array are built respectively, which can be used to estimate the space-conical angles, and the intersecting line of two half-conical surfaces is the path of arrival. With this method, training set is proved to be largely reduced, and also the model-building difficulty is effectively depressed. At last, simulations show this method has very high precision, so it has a broad application value.
同期刊论文项目
同项目期刊论文
期刊信息
位置：