基于神经网络的雷达抗干扰方法综述

A survey of the radar ECCM methods based on neural network

针对目前雷达抗干扰方法存在的局限性,提出了将神经网络技术应用于雷达抗干扰的研究思路。首先介绍了在雷达抗干扰领域应用比较成功的3种神经网络模型:BP网络、RBF网络和Kohonen网络;然后概括了神经网络应用于雷达抗干扰的3种主要途径:神经网络用于建模、用于特征提取和用于分类;最后对下一步的研究进行了总结与展望,分析了神经网络与其他技术相结合在雷达抗干扰中的应用前景。

With the consideration of the limitation of current radar ECCM methods, a new idea of applying neural network to radar ECCM is proposed. Firstly, three kinds of neural network models （ BP network, RBF network and Kohonen network） that are applied successfully in radar ECCM are introduced. Then three main ways （ modeling, feature extracting and classifying） of applying neural net-work in radar ECCM are generalized. Finally the combination of the neural network and other intelligent technologies such as time frequency analysis and genetic algorithm are analyzed.