单脉冲雷达天线方向图的BP神经网络建模

Modeling of Antenna Patterns for Monopulse Radar with BP Neural Networks

为了提高单脉冲雷达仿真速度,提出了应用BP神经网络对单脉冲雷达天线方向图进行建模的方法,并根据这一方法建立了某型单脉冲雷达天线方向图的BP神经网络模型。通过对比BP神经网络建模得到的单脉冲雷达天线方向图和实际单脉冲雷达天线方向图,并考虑到较小的BP神经网络训练均方误差,验证了该方法的有效性。实际的某型单脉冲雷达系统仿真试验表明利用建立的BP神经网络模型能大大地提高雷达仿真速度。

To improve the velocity of monopulse radar simulation,the method of modeling the antenna patterns of monopulse radar with BP neural networks is put forward, and the antenna patterns of a monopulse radar are modeled based on the method. Through comparing the BP neural network models with the real antenna patterns of the monopulse radar and considering the small mean square error of BP neural network training, the effectiveness of the method is verified. The monopulse radar system simulation tests indicate that the velocity of radar simulation can be improved greatly by making use of the BP neural network models.