基于卷积神经网络的微弱雷达信号增强技术研究
|
摘要
针对雷达信号侦察领域中微弱信号检测困难的问题,提出了一种利用卷积神经网络算法增强微弱信号的方法。在Tensor Flow框架下,首先对信号做短时傅里叶变换,通过监督学习训练,在含噪频谱和纯净频谱之间建立关系,最后将训练成熟的网络用于提高微弱信号的信噪比,实现了信号增强的目的。仿真结果表明,算法对含噪信号的增强效果十分明显。该算法的实现将有效提高信号检测概率,为信号检测领域提供有效支撑。
In order to solve the the difficulty of weak signal detection in the field of radar signal reconnaissance,a convolution neural network algorithm is proposed to enhance weak signal. In the TensorFlow framework,the signal is firstly made short time Fourier transform,and the relationship between the noise spectrum and the pure spectrum is established by supervising the learning and training. Finally,the mature network is used to improve the signal to noise ratio of the weak signal,and the purpose of signal enhancement is realized. Simulation results show that the algorithm is very effective for enhancing noisy signals. The implementation of the algorithm will effectively improve the probability of signal detection and provide effective support for signal detection.
In order to solve the the difficulty of weak signal detection in the field of radar signal reconnaissance,a convolution neural network algorithm is proposed to enhance weak signal. In the TensorFlow framework,the signal is firstly made short time Fourier transform,and the relationship between the noise spectrum and the pure spectrum is established by supervising the learning and training. Finally,the mature network is used to improve the signal to noise ratio of the weak signal,and the purpose of signal enhancement is realized. Simulation results show that the algorithm is very effective for enhancing noisy signals. The implementation of the algorithm will effectively improve the probability of signal detection and provide effective support for signal detection.
引文
[1]赵凯凯,杨承志,王龙.基于WCPF和FRFT的LPI信号识别研究[J].现代防御技术,2016,44(3):186-193+207.
[2] Montiel Zafra V,Canadas Quesada F J,Vera Candeas P,et al. A novel method to remove GPR background noise based on the similarity of non-neighboring regions[J].Journal of Applied Geophysics,2017,144:188-203.
[3]张梦.激光雷达回波信号增强技术研究[J].激光杂志,2017,38(9):146-149.
[4]谢锡海,魏晓会.一种FMCW雷达信号的小波阈值函数去噪法[J].西安邮电大学学报,2017,22(4):21-25.
[5]郭凤霞,戚俊,陈斐楠,等.基于小波变换的声雷达模拟信号去噪研究[J].现代雷达,2016,38(3):86-90.
[6]李立.模糊聚类在雷达频率信号目标检测处理中的应用[J].舰船科学技术,2017(6):189-191.
[7]刘帆,刘鹏远,李兵,等. Tensor Flow平台下的视频目标跟踪深度学习模型设计[J].激光与光电子学进展,2017,54(9):277-285.
[8] Abadi M,Agarwal A,Barham P,et al. Tensor Flow:large-scale machine learning on heterogeneous distributed systems[EB/OL].[2016]. https://arxiv. org/abs/1603. 04467.
[9]周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.
[10]王华利,邹俊忠,张见,等.基于深度卷积神经网络的快速图像分类算法[J].计算机工程与应用,2017,53(13):181-188.
[11] Glorot X,Bengio Y. Understanding the difficulty of training deep feedforward neural networks[J]. Journal of Machine Learning Research,2010,9:249-256.
[12] Le Q V,Jaitly N,Hinton G E. A simple way to initialize recurrent networks of rectified linear units[J]. Computer Science,2015.
[13] Kingma D P,Ba J. Adam:A method for stochastic optimization[EB/OL].[2014]. https://arxiv. org/abs/1412. 6980.
[14]镇凡迪,李倩,方烜.基于经验模态分解的自适应模态极大值去噪方法[J].通信电源技术,2016,33(5):80-82.
[15]马晓婷,尚宇. ECG去噪算法的设计与实现[J].国外电子测量技术,2017,36(7):18-22.
[16]肖小兵,刘宏立,马子骥.基于奇异谱分析的经验模态分解去噪方法[J].计算机工程与科学,2017,39(5):919-924.
[2] Montiel Zafra V,Canadas Quesada F J,Vera Candeas P,et al. A novel method to remove GPR background noise based on the similarity of non-neighboring regions[J].Journal of Applied Geophysics,2017,144:188-203.
[3]张梦.激光雷达回波信号增强技术研究[J].激光杂志,2017,38(9):146-149.
[4]谢锡海,魏晓会.一种FMCW雷达信号的小波阈值函数去噪法[J].西安邮电大学学报,2017,22(4):21-25.
[5]郭凤霞,戚俊,陈斐楠,等.基于小波变换的声雷达模拟信号去噪研究[J].现代雷达,2016,38(3):86-90.
[6]李立.模糊聚类在雷达频率信号目标检测处理中的应用[J].舰船科学技术,2017(6):189-191.
[7]刘帆,刘鹏远,李兵,等. Tensor Flow平台下的视频目标跟踪深度学习模型设计[J].激光与光电子学进展,2017,54(9):277-285.
[8] Abadi M,Agarwal A,Barham P,et al. Tensor Flow:large-scale machine learning on heterogeneous distributed systems[EB/OL].[2016]. https://arxiv. org/abs/1603. 04467.
[9]周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.
[10]王华利,邹俊忠,张见,等.基于深度卷积神经网络的快速图像分类算法[J].计算机工程与应用,2017,53(13):181-188.
[11] Glorot X,Bengio Y. Understanding the difficulty of training deep feedforward neural networks[J]. Journal of Machine Learning Research,2010,9:249-256.
[12] Le Q V,Jaitly N,Hinton G E. A simple way to initialize recurrent networks of rectified linear units[J]. Computer Science,2015.
[13] Kingma D P,Ba J. Adam:A method for stochastic optimization[EB/OL].[2014]. https://arxiv. org/abs/1412. 6980.
[14]镇凡迪,李倩,方烜.基于经验模态分解的自适应模态极大值去噪方法[J].通信电源技术,2016,33(5):80-82.
[15]马晓婷,尚宇. ECG去噪算法的设计与实现[J].国外电子测量技术,2017,36(7):18-22.
[16]肖小兵,刘宏立,马子骥.基于奇异谱分析的经验模态分解去噪方法[J].计算机工程与科学,2017,39(5):919-924.