针对多雷达辐射源脉冲交错背景下,线性调频(Linear Frequency Modulation,LFM)信号低信噪比导致的脉冲分裂带来原始信号参数难以估计的问题,本文提出了基于深度神经网络和直方图统计的LFM信号两阶段提取与参数估计方法。首先利用双向长...针对多雷达辐射源脉冲交错背景下,线性调频(Linear Frequency Modulation,LFM)信号低信噪比导致的脉冲分裂带来原始信号参数难以估计的问题,本文提出了基于深度神经网络和直方图统计的LFM信号两阶段提取与参数估计方法。首先利用双向长短时记忆网络挖掘原始脉冲流中LFM信号与非LFM信号的调制模式差异并进行分类;其次通过序列调频斜率直方图寻找LFM信号分裂脉冲序列间隐含的原始信号调频斜率信息,提取不同调频斜率的LFM信号脉冲子序列;最后在每个子序列中分别估计原始信号的参数。仿真实验结果表明,相较于传统的序列差值直方图算法和循环神经网络分选方法,本文所提方法能够更准确地提取出LFM脉冲信号,并得到较为精确的参数估计结果。展开更多
To enhance the capacity of the radar-reconnaissance interception receiver recognizing linear frequency modulated (LFM) at a low signal-noise ratio, this paper presents WignerHough transform (WHT) of the LFM signal and...To enhance the capacity of the radar-reconnaissance interception receiver recognizing linear frequency modulated (LFM) at a low signal-noise ratio, this paper presents WignerHough transform (WHT) of the LFM signal and its corresponding characteristics, derives the probability density functions of the LFM signal and Gaussian white noise within WHT based on entropy (WHTE), dimension under different assumptions and puts forward a WHT algorithm based on entropy of slice to improve the capacity of detecting the LFM signal. Entropy of the WHT domain slice is adopted to assess the information size of polar radius or angle slice, which is converted into the weight factor to weight every slice. Double-deck weight is used to weaken the influences of noise and disturbance terms and WHTE treatment and signal detection procedure are also summarized. The rationality of the algorithm is demonstrated through theoretical analysis and formula derivation, the efficiency of the algorithm is verified by simulation comparison between WHT, fractional Fourier transform and periodic WHT, and it is highlighted that the WHTE algorithm has better detection accuracy and range of application against strong noise background.展开更多
A novel algorithm is proposed to solve the poor per- formance problem of the Tent chaos-based frequency modulation （FM） signal for range-Doppler imaging, which takes it into complex multi-segment system by increasin...A novel algorithm is proposed to solve the poor per- formance problem of the Tent chaos-based frequency modulation （FM） signal for range-Doppler imaging, which takes it into complex multi-segment system by increasing its segments. The simulation results show that the effectiveness of the proposed algorithm, as well as the performance of the improved Tent FM signal is obvious in a multipath or noise propagation environment.展开更多
The Radon-ambiguity transform （RAT）, although efficient for detecting the linear frequency modulated signals （LFMs）, is troubled by the energy accumulation of noise in low signal-to-noise ratio （SNR）. A secondor...The Radon-ambiguity transform （RAT）, although efficient for detecting the linear frequency modulated signals （LFMs）, is troubled by the energy accumulation of noise in low signal-to-noise ratio （SNR）. A secondorder difference （SOD） method is proposed to treat with this problem. In the SOD method, the optimal search step and difference step are derived from the LFM rate resolution formula. The sharpness of the peaks of RAT is measured by curvature, and the sharpness, but not the magnitude of the peaks, is used to detect the LFMs. The SOD method removes the noise energy accumulation and reserves the drastically changing components integrally; thus, it improves the detection probability of LFMs in low SNR. The expected performance of the new method is verified by 100 Monte Carlo simulations.展开更多
The instantaneous frequency （IF） estimation of the linear frequency modulated （LFM） signals with time-varying amplitude using the peak of the Wigner-Ville distribution （WVD） is studied. Theoretical analysis show...The instantaneous frequency （IF） estimation of the linear frequency modulated （LFM） signals with time-varying amplitude using the peak of the Wigner-Ville distribution （WVD） is studied. Theoretical analysis shows that the estimation on LFM signals with time-varying amplitude is unbiased, only if WVD of time-varying amplitude reaches its maximum at frequency zero no matter in which time. The statistical performance in the case of additive white Guassian noise is evaluated and an analytical expression for the variance is provided. The simulations using LFM signals with Gaussian envelope testify that IF can be estimated accurately using the peak of WVD for four models of amplitude variation. Furthermore the statistical result of estimation on the signals with amplitude descending before rising is better than that of the signals with constant amplitude when the amplitude variation rate is moderate.展开更多
In this paper,parameter estimation of linear frequency modulation(LFM)signals containing additive white Gaussian noise is studied.Because the center frequency estimation of an LFM signal is affected by the error propa...In this paper,parameter estimation of linear frequency modulation(LFM)signals containing additive white Gaussian noise is studied.Because the center frequency estimation of an LFM signal is affected by the error propagation effect,resulting in a higher signal to noise ratio(SNR)threshold,a parameter estimation method for LFM signals based on time reversal is proposed.The proposed method avoids SNR loss in the process of estimating the frequency,thus reducing the SNR threshold.The simulation results show that the threshold is reduced by 5 dB compared with the discrete polynomial transform(DPT)method,and the root-mean-square error(RMSE)of the proposed estimator is close to the Cramer-Rao lower bound(CRLB).展开更多
This paper presents a method on non-linear correction of broadband LFMCW signal utilizing its relative nonlinear error. The deriving procedure and the results simulated by a computer and tested by a practical system a...This paper presents a method on non-linear correction of broadband LFMCW signal utilizing its relative nonlinear error. The deriving procedure and the results simulated by a computer and tested by a practical system are also introduced. The method has two obvious advantages compared with the previous methods: (1) Correction has no relation with delay time td and sweep bandwidth B; (2) The inherent non-linear error of VCO has no influence on the correction and its last results.展开更多
为了完成线性调频(linear frequency modulation,LFM)信号的稀疏采样,并利用稀疏数据对原始信号参数进行估计,本文提出了一种基于Z变换和改进有限新息率(finite rate of innovation,FRI)的LFM信号参数估计方法。以Z变换理论为基础,设计...为了完成线性调频(linear frequency modulation,LFM)信号的稀疏采样,并利用稀疏数据对原始信号参数进行估计,本文提出了一种基于Z变换和改进有限新息率(finite rate of innovation,FRI)的LFM信号参数估计方法。以Z变换理论为基础,设计了一种数学模型,一旦信号能够表达成该数学模型的结构形式,就能通过Z变换和零化滤波器的方法估计信号参数。然后,利用了自相关延迟的FRI结构对LFM信号采样,该结构不仅完成了LFM信号的稀疏采样,而且稀疏采样结果能够与数学模型结构相符。在理论上通过数学论证的方式证明了所提方法能够用于获取LFM信号参数信息,并通过仿真和实测数据验证了所提方法的有效性,理论和实验结果表明该方法只需要4个采样点就能实现对LFM信号的参数估计,并且实验中的参数估计误差均在3%以内,极大的提高有限新息率采样的参数估计效率。展开更多
在探讨正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统的优化中,一个显著挑战在于其信号检测性能的相对不足。同时,针对基于深度神经网络的索引调制(Deep Neural Network Based Index Modulation,DNN-IM)检测算法...在探讨正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统的优化中,一个显著挑战在于其信号检测性能的相对不足。同时,针对基于深度神经网络的索引调制(Deep Neural Network Based Index Modulation,DNN-IM)检测算法,普遍存在着误码率及损失值偏高的问题。为了弥补上述难题,文中提出一种基于多层感知机(Multilayer Perceptron,MLP)的索引调制检测算法,即MLP-IM算法。该算法采用融合两个连接层与一个输出层的架构设计,通过挑选的激活函数实现对OFDM索引调制系统中数据比特的精准还原。首先将OFDM索引调制系统的基础理论巧妙应用于数据的预处理阶段,随后利用仿真数据集对MLP神经网络模型进行全面而深入的离线训练,确保模型的稳健性与准确性。在检测阶段,通过MLP-IM检测算法实现了对OFDM索引调制系统的高效检测。仿真结果表明,所提出的MLP-IM算法在误码率控制和损失值两个方面的性能表现与最大似然检测算法相媲美,甚至在某些场景下超越了现有DNN-IM算法的性能,其性能改善幅度在0.2~6 dB的区间内。展开更多
文摘针对多雷达辐射源脉冲交错背景下,线性调频(Linear Frequency Modulation,LFM)信号低信噪比导致的脉冲分裂带来原始信号参数难以估计的问题,本文提出了基于深度神经网络和直方图统计的LFM信号两阶段提取与参数估计方法。首先利用双向长短时记忆网络挖掘原始脉冲流中LFM信号与非LFM信号的调制模式差异并进行分类;其次通过序列调频斜率直方图寻找LFM信号分裂脉冲序列间隐含的原始信号调频斜率信息,提取不同调频斜率的LFM信号脉冲子序列;最后在每个子序列中分别估计原始信号的参数。仿真实验结果表明,相较于传统的序列差值直方图算法和循环神经网络分选方法,本文所提方法能够更准确地提取出LFM脉冲信号,并得到较为精确的参数估计结果。
基金supported by the Aeronautical Science Fund of China(201455960252015209619)
文摘To enhance the capacity of the radar-reconnaissance interception receiver recognizing linear frequency modulated (LFM) at a low signal-noise ratio, this paper presents WignerHough transform (WHT) of the LFM signal and its corresponding characteristics, derives the probability density functions of the LFM signal and Gaussian white noise within WHT based on entropy (WHTE), dimension under different assumptions and puts forward a WHT algorithm based on entropy of slice to improve the capacity of detecting the LFM signal. Entropy of the WHT domain slice is adopted to assess the information size of polar radius or angle slice, which is converted into the weight factor to weight every slice. Double-deck weight is used to weaken the influences of noise and disturbance terms and WHTE treatment and signal detection procedure are also summarized. The rationality of the algorithm is demonstrated through theoretical analysis and formula derivation, the efficiency of the algorithm is verified by simulation comparison between WHT, fractional Fourier transform and periodic WHT, and it is highlighted that the WHTE algorithm has better detection accuracy and range of application against strong noise background.
基金supported by the National Natural Science Foundation of China (610320106110117211076006)
文摘A novel algorithm is proposed to solve the poor per- formance problem of the Tent chaos-based frequency modulation （FM） signal for range-Doppler imaging, which takes it into complex multi-segment system by increasing its segments. The simulation results show that the effectiveness of the proposed algorithm, as well as the performance of the improved Tent FM signal is obvious in a multipath or noise propagation environment.
基金supported by the Program for New Century Excellent Talents in University, Ministry of Education (NCET-05-0803)
文摘The Radon-ambiguity transform （RAT）, although efficient for detecting the linear frequency modulated signals （LFMs）, is troubled by the energy accumulation of noise in low signal-to-noise ratio （SNR）. A secondorder difference （SOD） method is proposed to treat with this problem. In the SOD method, the optimal search step and difference step are derived from the LFM rate resolution formula. The sharpness of the peaks of RAT is measured by curvature, and the sharpness, but not the magnitude of the peaks, is used to detect the LFMs. The SOD method removes the noise energy accumulation and reserves the drastically changing components integrally; thus, it improves the detection probability of LFMs in low SNR. The expected performance of the new method is verified by 100 Monte Carlo simulations.
文摘The instantaneous frequency （IF） estimation of the linear frequency modulated （LFM） signals with time-varying amplitude using the peak of the Wigner-Ville distribution （WVD） is studied. Theoretical analysis shows that the estimation on LFM signals with time-varying amplitude is unbiased, only if WVD of time-varying amplitude reaches its maximum at frequency zero no matter in which time. The statistical performance in the case of additive white Guassian noise is evaluated and an analytical expression for the variance is provided. The simulations using LFM signals with Gaussian envelope testify that IF can be estimated accurately using the peak of WVD for four models of amplitude variation. Furthermore the statistical result of estimation on the signals with amplitude descending before rising is better than that of the signals with constant amplitude when the amplitude variation rate is moderate.
基金supported by the Regional Joint Fund for Basic and Applied Basic Research of Guangdong Province(2019B1515120009)the Defense Basic Scientific Research Program(61424132005).
文摘In this paper,parameter estimation of linear frequency modulation(LFM)signals containing additive white Gaussian noise is studied.Because the center frequency estimation of an LFM signal is affected by the error propagation effect,resulting in a higher signal to noise ratio(SNR)threshold,a parameter estimation method for LFM signals based on time reversal is proposed.The proposed method avoids SNR loss in the process of estimating the frequency,thus reducing the SNR threshold.The simulation results show that the threshold is reduced by 5 dB compared with the discrete polynomial transform(DPT)method,and the root-mean-square error(RMSE)of the proposed estimator is close to the Cramer-Rao lower bound(CRLB).
文摘This paper presents a method on non-linear correction of broadband LFMCW signal utilizing its relative nonlinear error. The deriving procedure and the results simulated by a computer and tested by a practical system are also introduced. The method has two obvious advantages compared with the previous methods: (1) Correction has no relation with delay time td and sweep bandwidth B; (2) The inherent non-linear error of VCO has no influence on the correction and its last results.
文摘为了完成线性调频(linear frequency modulation,LFM)信号的稀疏采样,并利用稀疏数据对原始信号参数进行估计,本文提出了一种基于Z变换和改进有限新息率(finite rate of innovation,FRI)的LFM信号参数估计方法。以Z变换理论为基础,设计了一种数学模型,一旦信号能够表达成该数学模型的结构形式,就能通过Z变换和零化滤波器的方法估计信号参数。然后,利用了自相关延迟的FRI结构对LFM信号采样,该结构不仅完成了LFM信号的稀疏采样,而且稀疏采样结果能够与数学模型结构相符。在理论上通过数学论证的方式证明了所提方法能够用于获取LFM信号参数信息,并通过仿真和实测数据验证了所提方法的有效性,理论和实验结果表明该方法只需要4个采样点就能实现对LFM信号的参数估计,并且实验中的参数估计误差均在3%以内,极大的提高有限新息率采样的参数估计效率。
文摘在探讨正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统的优化中,一个显著挑战在于其信号检测性能的相对不足。同时,针对基于深度神经网络的索引调制(Deep Neural Network Based Index Modulation,DNN-IM)检测算法,普遍存在着误码率及损失值偏高的问题。为了弥补上述难题,文中提出一种基于多层感知机(Multilayer Perceptron,MLP)的索引调制检测算法,即MLP-IM算法。该算法采用融合两个连接层与一个输出层的架构设计,通过挑选的激活函数实现对OFDM索引调制系统中数据比特的精准还原。首先将OFDM索引调制系统的基础理论巧妙应用于数据的预处理阶段,随后利用仿真数据集对MLP神经网络模型进行全面而深入的离线训练,确保模型的稳健性与准确性。在检测阶段,通过MLP-IM检测算法实现了对OFDM索引调制系统的高效检测。仿真结果表明,所提出的MLP-IM算法在误码率控制和损失值两个方面的性能表现与最大似然检测算法相媲美,甚至在某些场景下超越了现有DNN-IM算法的性能,其性能改善幅度在0.2~6 dB的区间内。
