基于深度神经网络模型的雷达目标识别

Radar Target Recogintion Based on Deep Neural Network

根据雷达测量的目标电磁散射面积（RCS）序列，采用深度神经网络模型识别空间飞行目标。首先，阐述了提取RCS时间序列特征的方法，包括均值、均方差及周期特性等特征；然后，给出了深度神经网络模型识别RCS目标的算法；最后，用仿真数据验证该识别方法，数值实验结果表明该方法能较准确识别雷达跟踪目标。

The deep neral network for space flying target recogniton is provided using radar cross section （RCS） . Firstly, the methed of extracting RCS time series characteristics is described, including the average, mean variance and periodicity chracteristries. Then, a deep neural network model is presented to identify RCS targets. Finally, the identification method is verified by simulation data, and the numerical results show that the method can accurately identify radar tracking targets.