In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intr...In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intra-pulse modulation signal type based on deep residual network.The basic principle of the recognition method is to obtain the transformation relationship between the time and frequency of complex radar intra-pulse modulation signal through short-time Fourier transform(STFT),and then design an appropriate deep residual network to extract the features of the time-frequency map and complete a variety of complex intra-pulse modulation signal type recognition.In addition,in order to improve the generalization ability of the proposed method,label smoothing and L2 regularization are introduced.The simulation results show that the proposed method has a recognition accuracy of more than 95%for complex radar intra-pulse modulation sig-nal types under low SNR(2 dB).展开更多
This paper presents a joint high order statistics （HOS） and signal-to-noise ratio （SNR） algorithm for the recognition of multiple-input multiple-output （MIMO） radar signal without a priori knowledge of the signa...This paper presents a joint high order statistics （HOS） and signal-to-noise ratio （SNR） algorithm for the recognition of multiple-input multiple-output （MIMO） radar signal without a priori knowledge of the signal parameters. This method is capable of recognizing the MIMO radar signal as well as discriminating it from single-carrier signal adopted by conventional radar. Meanwhile, the sub-carrier number of the none-coding MIMO radar signal is estimated. Extensive simulations are carried out in different operating conditions. Simulation results prove the feasibility and indicate that the recognition probability could reach over 90% when the value of SNR is above 0 dB.展开更多
针对已有Cohen类时频分布等方法时频聚焦能力不足、在低信噪比(signal to noise ratio,SNR)情况下调制识别准确率低的问题,提出一种基于同步提取变换(synchro-extracting transform,SET)去噪的分组卷积神经网络调制识别方法。所提方法使...针对已有Cohen类时频分布等方法时频聚焦能力不足、在低信噪比(signal to noise ratio,SNR)情况下调制识别准确率低的问题,提出一种基于同步提取变换(synchro-extracting transform,SET)去噪的分组卷积神经网络调制识别方法。所提方法使用SET对雷达信号进行时频分析,以获得良好的时频聚焦性,提高时频分析的计算效率;通过Viterbi算法搜索估计时频系数矩阵中的瞬时频率轨迹,综合考虑信号能量强度分布与瞬时频率轨迹的平滑性,并对得到的瞬时频率轨迹进行中值滤波以去除脉冲噪声;保留瞬时频率轨迹邻域的时频系数,以达到时频图去噪的目的。最后,将去噪后的时频图送入具有残差连接的分组卷积神经网络进行特征提取与调制识别。实验结果表明,当SNR为-12 dB时,去噪后的SET时频图时频聚焦性好,调制识别准确率比未去噪的识别准确率提高了13.69%,证明所提出的雷达信号调制识别方法在低SNR条件下对多种复杂调制类型的信号具有良好的识别性能。展开更多
低截获概率(low probability of intercept,LPI)雷达已成为新时代雷达装备中关键的技术体制或工作模式,针对LPI雷达信号调制识别及参数估计方法的研究是当前雷达对抗侦察领域的热点。首先,分析了几种典型LPI雷达信号的脉内特征,梳理了LP...低截获概率(low probability of intercept,LPI)雷达已成为新时代雷达装备中关键的技术体制或工作模式,针对LPI雷达信号调制识别及参数估计方法的研究是当前雷达对抗侦察领域的热点。首先,分析了几种典型LPI雷达信号的脉内特征,梳理了LPI雷达信号调制识别及参数估计的传统和主流方法,并说明其原理、优缺点和研究现状。最后,总结了现有LPI雷达信号调制识别及参数估计方法尚存的问题,并指出其未来发展趋势,旨在为今后的研究提供参考。展开更多
本文针对低截获概率(Low Probability of Intercept,LPI)雷达信号调制类型的识别问题提出了一种基于Swin Transformer神经网络的识别方法.该方法首先用平滑伪Wigner-Ville分布对信号进行时频变换,将一维时域信号转换为二维时频图像,然...本文针对低截获概率(Low Probability of Intercept,LPI)雷达信号调制类型的识别问题提出了一种基于Swin Transformer神经网络的识别方法.该方法首先用平滑伪Wigner-Ville分布对信号进行时频变换,将一维时域信号转换为二维时频图像,然后使用Swin Transformer神经网络对时频图像进行特征提取和调制类型识别.仿真结果显示该方法具有较强的抗噪声能力,在低信噪比条件下识别准确率高,且具有较强的小样本适应能力.展开更多
针对低信噪比条件下复杂多类雷达信号调制方式识别率低的问题,本文提出了一种基于时频分析和深度学习的雷达信号调制方式识别方法.利用CTFD(Cohen class Time-Frequency Distribution)时频分析将信号时域波形变换为二维时频图像,更清晰...针对低信噪比条件下复杂多类雷达信号调制方式识别率低的问题,本文提出了一种基于时频分析和深度学习的雷达信号调制方式识别方法.利用CTFD(Cohen class Time-Frequency Distribution)时频分析将信号时域波形变换为二维时频图像,更清晰地表征信号特征;采用灰度化和双三次插值运算等方法对时频图预处理,实现图像通道数和尺寸的减少,以降低深度学习模型数据输入量;进一步调整输入输出通道数构建小型EfficientNet网络,再由多个小型网络并行处理构建分裂网络EfficientNet-B0-Split3,将时频图像输入网络实现雷达信号调制方式识别.实验结果表明,在信噪比为-8 dB时,新方法对17类不同调制方式的雷达信号整体识别率可达97.1%,相对于扩张残差网络提高约2.4个百分点;在信噪比为-10 dB时,识别率可达92.1%,相对于EfficientNet提高约0.7个百分点,提升了低信噪比条件下复杂多类雷达信号调制方式识别率.展开更多
文摘In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intra-pulse modulation signal type based on deep residual network.The basic principle of the recognition method is to obtain the transformation relationship between the time and frequency of complex radar intra-pulse modulation signal through short-time Fourier transform(STFT),and then design an appropriate deep residual network to extract the features of the time-frequency map and complete a variety of complex intra-pulse modulation signal type recognition.In addition,in order to improve the generalization ability of the proposed method,label smoothing and L2 regularization are introduced.The simulation results show that the proposed method has a recognition accuracy of more than 95%for complex radar intra-pulse modulation sig-nal types under low SNR(2 dB).
基金supported by the Foundation of Chinese People’s Liberation Army General Equipment Department(41101020303)
文摘This paper presents a joint high order statistics （HOS） and signal-to-noise ratio （SNR） algorithm for the recognition of multiple-input multiple-output （MIMO） radar signal without a priori knowledge of the signal parameters. This method is capable of recognizing the MIMO radar signal as well as discriminating it from single-carrier signal adopted by conventional radar. Meanwhile, the sub-carrier number of the none-coding MIMO radar signal is estimated. Extensive simulations are carried out in different operating conditions. Simulation results prove the feasibility and indicate that the recognition probability could reach over 90% when the value of SNR is above 0 dB.
文摘低截获概率(low probability of intercept,LPI)雷达已成为新时代雷达装备中关键的技术体制或工作模式,针对LPI雷达信号调制识别及参数估计方法的研究是当前雷达对抗侦察领域的热点。首先,分析了几种典型LPI雷达信号的脉内特征,梳理了LPI雷达信号调制识别及参数估计的传统和主流方法,并说明其原理、优缺点和研究现状。最后,总结了现有LPI雷达信号调制识别及参数估计方法尚存的问题,并指出其未来发展趋势,旨在为今后的研究提供参考。
文摘本文针对低截获概率(Low Probability of Intercept,LPI)雷达信号调制类型的识别问题提出了一种基于Swin Transformer神经网络的识别方法.该方法首先用平滑伪Wigner-Ville分布对信号进行时频变换,将一维时域信号转换为二维时频图像,然后使用Swin Transformer神经网络对时频图像进行特征提取和调制类型识别.仿真结果显示该方法具有较强的抗噪声能力,在低信噪比条件下识别准确率高,且具有较强的小样本适应能力.
