Parameters Estimation For Cyclostationary Signals In Stable Impulsive Noise

There are strong impulsive noise in communications,oceans,astronomy and other areas.Many of the signals we encountered in these systems are cyclostationary signal,which is a special kind of non-stationary signal both with periodic or multi-periodic stable transform statistics.The tradition cyclic stationary signal DOA estimation algorithm is using this kind of cyclic stationary characteristic to estimate the DOA.In fact,there are many non-Gaussian signals and noises with notable pulse characteristic,so people usually use alpha-stable distribution model to distribution model to depict stochastic signals with a remarkable impulsive characteristic.Because of the second-order statistics of the alpha-stable distributed noise doesn't exist,which result in noticeable degradation or even failure of the performance of algorithms based on the second-order statistics.Focusing on this question,scholars will second-order cyclic statistics(CCO)is extended to the fractional lower lower order statistics(FLOCC),is proposed based on DOA estimation algorithm of FLOCC.But this algorithm needs known impulse noise prior knowledge in general is hard to get in actual application.Therefore,this thesis defines the concept of generalized correntropy cyclic statistics(CECCO)and nonlinear transform-based cyclic correlation statistics(NTCCO),and its application in the far-field under the a-stable distribution of signal DOA estimation algorithm,this thesis presents respectively the algorithms of Direction of Arrival(DOA)estimation based on CECCO and NTCCO.Stimulation results show that the proposed two algorithms can effectually give DOA estimation under impulsive noise conditions,and they performance is superior to the CCO-MUSIC based on second order cyclic correlation and FLOCC-MUSIC based on fractional lower order cyclic correlation.Especially in the case of low signal to noise ratio,the performance of the proposed algorithm is particularly prominent.Secondly,since the source located in Fresnel area relative to the array aperture,the far field assumption can not be used as array receives spherical waves.So the corresponding algorithms will fail and the study for the near field source will be more complicated.To solve this problem,the FLOCC,CECCO and NTCCO are combined with near-field model,and new DOA and range estimation methods for the near-field cyclic stationary signal which applied to alpha-stable distribution noise environment are proposed.Finally,the signal of the airspace sparse feature combining FLOCC,CECCO and NTCCO statistics,proposed based on the sparse reconstruction the near field of the theory of cyclic stationary signal DOA and distance estimation in the stable distribution of impulsive noise environment.The algorithm is compared with the subspace algorithm with high resolution,without a priori knowledge about the source number,and many other advantages.