摘要
提出一种使用天牛须搜索(beetle antennae search,BAS)算法优化反向传播(back propagation,BP)神经网络以识别遮盖干扰信号的方法。采用BAS优化BP神经网络的初始权值、阈值,以适应度函数作为评价标准,优化出最佳的权值、阈值,训练BP神经网络,得到最优BP神经网络模型,使用优化后的BP神经网络对雷达有源遮盖性干扰信号进行分类识别。选取射频噪声、噪声调幅和噪声调频3种干扰信号进行仿真,结果表明,BAS-BP神经网络和BP神经网络的均方误差分别为0.1486和0.1770,平均绝对值误差分别为0.2197和0.2693。BAS-BP神经网络和BP神经网络对3种干扰信号的平均识别率分别为0.9137和0.8827。BAS-BP神经网络方法能够识别干扰信号,且效果优于BP神经网络算法。
A new back propagation(BP)neural network mode based on beetle antennae search(BAS)algorithm is proposed to classify the radar covering jamming signal.BAS is used to optimize the initial weights and thresholds of the neural network.Using the fitness function as the evaluation criteria,the optimal weights and thresholds is obtained.After training the neural network again,the optimal network model is then obtained.This optimized BP neural network can be used to classify the radar active covering jamming signals better.Three covering jamming signals,RF noise,noise AM and noise FM are chosen for simulation experiments.Results show that the mean square error of the BAS-BP neural network and the BP neural network are 0.1486 and 0.1770 respectively,the average absolute error is 0.2197 and 0.2693 respectively,and the average recognition rate of the BAS-BP neural network and the BP neural network under different jamming signals is 0.9137 and 0.8827 respectively.This optimized BP neural network can be used to classify the radar active covering jamming signals,and the effect is better than that of BP neural network algorithm.
作者
杨洁
褚书培
YANG Jie;CHU Shupei(School of Communication and Information Engineering,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)
出处
《西安邮电大学学报》
2019年第5期6-10,共5页
Journal of Xi’an University of Posts and Telecommunications
基金
陕西省教育厅专项科研计划资助项目(17JK0693)
关键词
遮盖性干扰信号
反向传播神经网络
天牛须搜索算法
covering jamming signal
back propagation neural network
beetle antennae search algorithm
