Parameter Estimation And Modulation Recognition Of Wireless Source In Complex Electromagnetic Environment

Parameter estimation and modulation recognition of wireless source in complex electromagnetic environment are central topics of signal processing for decades,and they are widely used in radio monitoring,military,communication and other fields.With the development of radio technology,the electromagnetic environment in wireless channel is becoming more and more complex,and the important frequency band is attacked maliciously or interfered unintentionally,which brings many challenges to radio management and monitoring.Under the current complicated electromagnetic environment,the traditional signal processing methods cannot works well under impulsive noise and co-channel interference.Therefore,it is urgent to develop new signal processing technology to solve the problems in radio management and monitoring under complex electromagnetic environmentTo fulfill the problems in radio management and monitoring at present,this thesis choses the signal modulation recognition and the interference localization in radio monitoring as the study background.Based on traditional algorithms,combined with the theory of cyclic correntropy and bounded nonlinear function,this thesis mainly study the carrier frequency estimation,time delay estimation and signal modulation recognition.Carrier frequency estimation and time delay estimation methods for impulsive noise and co-channel interference,as well as modulation recognition methods for impulsive noise are proposed.These methods provide new signal processing techniques for radio monitoring in complicated electromagnetic environments.The detailed research contents are listed as follows:(1)According to the performance degradation of the carrier frequency estimation method of time-frequency overlapped BPSK signals under impulsive random noise jamming,a novel carrier frequency estimation algorithm of the time-frequency overlapped BPSK signals based on the cyclic correntropy spectrum is proposed.The expression of cyclic correntropy spectrum of BPSK signal at zero frequency is derived.It is proved that the maximum value of cyclic correntropy spectrum at zero frequency can be obtained when the cyclic frequency is equal to twice of the signal carrier frequency,and the carrier frequency can be estimated by estimating the cyclic frequency of each time-frequency overlapped signal in the cyclic frequency domain.The proposed method can suppress the impulsive noise and has better estimation performance than that of the contrastive methods.Simulation results are presented to demonstrate the effectiveness and robustness of the proposed method.(2)According to the performance degradation of traditional cyclic delay estimation methods under the interference of impulsive random noise and co-channel interference,a generalized cyclic correntropy function is proposed by combining the correntropy-based correlation function with the cyclostationary property inspired by the cyclic correntropy function.A novel selective time delay estimation method based on generalized cyclic correntropy function is proposed,which is improved by replacing the cyclic correlation function with the generalized cyclic correntropy function.Simulation results show that the proposed method can effectively suppress the effects of impulsive random noise,Gaussian noise and co-channel interference,and is a robust selectivity delay estimation method.(3)According to the performance degradation of the traditional MPSK signal modulation recognition method based on high order cumulant under impulsive random noise,a bounded nonlinear function based on the hyperbolic tangent function(tanh)is proposed.The outliers in impulsive noise can be effectively suppressed by adjusting the parameters of the bounded nonlinear function to preprocess the signal.Three generalized higher-order cumulant classification feature parameters based on bounded nonlinear function are constructed.According to the three parameters,a modulation recognition method based on bounded function is proposed.The modulation mode classification of {BPSK,QPSK,OQPSK,8PSK} signals is realized.The proposed method can be effectively identified the signal modulation mode when ? ?[0.4,2],the accuracy close to 100%,which is a robust method.