| 摘要: |
| 针对传统方法在低信噪比情况下雷达信号特征提取困难、识别准确率较低的问题，提出了一种基于时频分析和深度学习的雷达信号识别方法。首先利用Choi-Williams分布将时域信号转换成时频图像，然后将时频图像作为网络的输入，通过弱化与增强残差块来实现对时频图像中噪声信息的弱化以及不同特征形态间差异性的增强，最终实现分类识别。实验结果表明，在信噪比为-10 dB情况下平均识别准确率仍能达到94.5%以上。 |
| 关键词: 雷达信号识别 低信噪比信号 时频分析 深度学习 弱化与增强 |
| DOI：10.20079/j.issn.1001-893x.231101001 |
|
| 基金项目:广西自然科学基金重点项目（2020GXNSFDA238001); |
|
| Radar Signal Recognition Based on Weakening and Strengthening Network |
| SHI Liquan,ZHANG Hongmei |
| (1.School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China;2.The Ministry of Education Key Laboratory of Cognitive Radio and Information Processing,Guilin 541004,China) |
| Abstract: |
| In response to the difficulties in extracting radar signal features and low recognition accuracy in low signal-to-noise ratio(SNR) situations using traditional methods,a radar signal recognition method based on time-frequency analysis and deep learning is proposed.Firstly,Choi-Williams distribution is used to convert the time-domain signal into a time-frequency image.Then,the time-frequency image is used as input to the network,and the noise information in the time-frequency image is weakened and the differences between different feature forms are enhanced through weakening and strengthening residual module,ultimately achieving classification recognition.The experimental results show that the average recognition accuracy can still reach over 94.5% even under a SNR of -10 dB. |
| Key words: radar signal recognition low signal-to-noise ratio signal time-frequency analysis deep learning weakening and strengthening |
