深度学习在雷达中的研究综述

雷达通过发射天线发射电磁波，经过不同物体反射接收到相应的反射波，对其接收结果进行分析，能得到物体距雷达的位置，径向运动速度等信息，所以对雷达信号的分析具有重要的研究意义。近些年深度学习成为各个领域的研究热点，而在雷达领域同样可通过深度学习算法实现对信号的相应的信息处理。与传统方法相比，深度学习算法具有自动提取深层特征、获取较高准确率等优势。该文具体介绍了近期典型的深度学习算法在雷达信号处理中的应用及研究情况。此外，该文介绍了两个在雷达领域中应用深度学习亟待解决的问题，即过拟合和可解译性。 

Study on Deep Learning in Radar

Electromagnetic waves are transmitted by radars and reflected by different objects, and radar signal processing is highly significant as its analyses can lead to the acquisition of important information such as the situation and radial movement speed. Moreover, deep learning has gained much attention in several fields, and it can be utilized to implement radar signal processing. Compared with the traditional methods, deep learning can realize automatic feature extraction and yield highly accurate results; hence, in this paper, the application of deep learning algorithm in radar signal processing is studied. In addition, the study directions in radar signal processing are summarized into overfitting and interpretation. Thus, these two issues are being considered.