基于改进型AlexNet的LPI雷达信号识别  

LPI radar signal recognition based on improved AlexNet

作　　者：杨洁 张欢 YANG Jie;ZHANG Huan(School of Communication and Information Engineering,Xi’an University of Posts and Telecommunications,Xi’an 710121,China)

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710121

出　　处：《现代电子技术》2020年第5期57-60,共4页Modern Electronics Technique

基　　金：国家自然科学基金青年科学基金项目(61402365);陕西省教育厅专项基金项目(17JK0693)。

摘　　要：为了识别低截获概率(LPI)雷达信号,给出一种基于Choi-Williams分布(CWD)和改进型AlexNet网络模型。首先,利用CWD时频分析方法获得LPI雷达信号的二维时频图像;然后,对获取的原始图像进行预处理,建立改进型AlexNet网络模型对处理后的图像进行训练,获得训练模型;最后,利用训练模型对常见LPI雷达信号(FMCW,Costas,Frank,P1,P2,P3,P4)进行识别。仿真结果表明,与AlexNet网络模型相比,改进型AlexNet对LPI雷达信号识别率更高。A Choi-Williams distribution(CWD)and an improved AlexNet network model are presented to identify the low probability of intercept(LPI)radar signals.Firstly,the two-dimensional time-frequency image of LPI radar signals is obtained by CWD time-frequency analysis method,and then the obtained original image is preprocessed.The improved AlexNet network model is created to train the processed image for the training model.Finally,the training model is used to identify common LPI radar signals(FMCW,Costas,Frank,P1,P2,P3,P4).The simulation results show that the improved AlexNet has a higher recognition rate for LPI radar signals than that of the AlexNet network model.

关 键 词：LPI雷达信号 Choi-Williams分布 时频图像 图像处理 深度学习 AlexNet 

分 类 号：TN957.52-34[电子电信—信号与信息处理]