Study On Theory And Method For Signal Processing Of EES-MIMO Radar With Impulsive Noise

Electromagnetic Environmental Sensory MIMO(EES-MIMO,Electromagnetic Environmental Sensory-Multiple Input Multiple Output)radar is MIMO radar with new mechanism;it improves the anti-interference ability of the MIMO radar by sensing the frequency characteristics of the space electromagnetic environment and changing the transmiting waveform according to the frequency characteristics.Traditional radar signal processing methods assume that the noise obeys the Gaussian distribution,but in the actual working environment of radar,there is not only Gaussian noise,but also a large number of non-Gaussian noises.Among them,the most serious influence on the existing signal processing methods comes from a class of non-Gaussian noise with impulsive property.A lot of research shows that this kind of impulsive noise can be described by a stable distribution.At present,How to effectively overcome and suppress the impact of such impulsive noise on the DOA estimation of MIMO radar has become an urgent problem to be solved in the field of signal processing.Based on the EES-MIMO radar signal processing problem in the background of impulsive noise,this paper focuses on the parameter estimation of ? stable distribution,the AR model spectral estimation of stable distribution,the DOA estimation method of EES-MIMO radar in a stable distribution noise background.Aiming at parameter estimation of a stable distribution,this paper presents a character exponential estimation method based on combinatorial distribution,a symmetric coefficient estimation method based on extreme characteristic and a dispersion coefficient estimation method based on probability distribution function.The character exponential estimation method based on the combined distribution is used to estimate the feature exponent by the relationship between the superposition result and the lower order moment of the original data fraction by using the linear addition characteristic and the p norm property of the ? stable distribution;the symmetric coefficient estimation method based on extreme characteristic estimates the symmetry coefficient using the trailing property of the ? stable distribution;the dispersion coefficient estimation method based on probability distribution function uses the probability distribution of the probability distribution function of the stable distribution to estimate the dispersion coefficient of the stable distribution.The above methods have improved the accuracy of the ? stable distribution parameter estimation,which have laid a good foundation for the research of EES-MIMO radar signal processing under the a stable distribution noise background.In the aspect of electromagnetic environmental sensory,this paper proposes an AR model covariance spectrum estimation method based on mutual covariance and minimum mutual square error for the problem of covariance spectrum estimation in the background of a stable distribution noise.The mutual covariance and minimum mutual square error method can overcome the channel superimposed noise by using the independent properties of the inter-channel noise,thus effectively improving the performance of the spatial AR spectrum estimation.Aiming at the DOA estimation problem of EES-MIMO radar in the background of? stable distribution noise,this paper mainly proposes the following EES-MIMO radar sources number and DOA estimation methods:1.The estimation of the sources number is the basis and premise of DOA estimation.Based on SCM(Shift Covariance Matrix)-Eigenvalue method,SCM-Gay circle method and SCM-Ostu inter-class variance method are used to estimate the number of sources.The Shift Covariance Matrix makes the covariance matrix overcome the influence of white a-stable noise on the diagonal elements of the covariance matrix and improve the SNR of the eigenvalue of the covariance matrix.Under the background of impulsive noise,the above three methods can effectively estimate the number of sources,which lays a good foundation for DOA estimation of EES-MIMO radar.2.This paper presents the DOA estimation method of SCM-Capon and SCM-MUSIC for EES-MIMO radar,when ?>1,these two methods can effectively estimate the DOA of the EES-MIMO radar.3.The concept of unity circle transformation(UCC,Unity Cir)is proposed.The unit circle transformation can not only keep the signal phase information,but also can effectively suppress the impulsive noise.The UCC-Capon and UCC-MUSIC methods of DOA estimation of EES-MIMO radar are proposed.The two methods do not need ? to know that the DOA of EES-MIMO radar narrowband signal can be effectively estimated in the range of 0<?<2.4.Some DOA estimation methods for EES-MIMO radar DOA estimation with colored impulsive noise is proposed.This method proposes a DOA estimation method based on cyclic fraction low order moment,cyclic fraction low order covariance and cyclic unit circle transform covariance.These methods effectively solve the DOA Estimation problem of EES-MIMO Radar in colored impulsive noise Background.5.In order to reduce the calculation of the DOA algorithm for spectral peak search,this paper proposes a DOA spectral peak search method based on sequential quadratic programming(SQP),and deduces the first derivative expression of DOA spectral search target function and analyzes the amount of calculation for the spectral peak search.The simulation results show that the DOA spectral peak search method based on SQP can effectively reduce the number of search points and the amount of computation and improve the efficiency of DOA estimation.