Research On Circular Array Two-dimensional Direction Finding Based On Coherent And Correlated Signals

As a key research content in array signal processing,direction of arrival(DOA)estimation has been diffusely used in radar positioning,wireless communication,sonar detection and medical fields.As a common array structure in passive radar direction finding system,uniform circular array(UCA)is compared with uniform linear array Uniform linear array(ULA),which can not only provide 360° azimuth information,but also has many advantages such as easy conformal and has conjugate symmetry of guidance vector,has attracted extensive attention and research in recent years.However,with the increasingly complex communication environment,the received signals of the array are no longer limited to simple independent signals,but also filled with a large number of related signals and even coherent signals.The coherent signal causes the rank deficiency of the covariance matrix of the array received data,and the conventional subspace algorithm can not be directly applied,a series of decorrelation pretreatment operations are needed to recover the rank of the covariance matrix,so as to make the subspace algorithm effective.Therefore,the study of circular array two-dimensional direction finding of coherent and correlated signals in the field of passive radar direction finding has not only theoretical significance,but also practical engineering value.This paper mainly studies some existing algorithms in the above problems and puts forward corresponding improvement schemes,mainly including the following two aspects:1.The incident problem of coherent signal based on uniform circular array is studied,and a two-dimensional DOA estimation method based on uniform circular array coherent signal source is proposed.The proposed method first uses a single circular array axial virtual translation technology to eliminate the interference of coherence,then denoising the data covariance matrix,combines the robustness of DOA matrix method and the eigendecomposition operation to estimate the elevation angle of the target signal.Then,using the beam conversion technology of the circular array,the smoothed covariance matrix is multiplied by the beam space transformation matrix to obtain the beam domain covariance matrix,so that the circular array becomes a virtual linear array with conjugate symmetry of the steering vector and Vandermonde structure.Finally,the azimuth of the target signal is estimated by the root MUSIC algorithm.The proposed method does not need complex two-dimensional spectral peak search,only obtains the angle information through a series of linear operations and engendecomposition,and the azimuth and elevation angles are matched automatically.The effectiveness and reliability of the method are verified by computer simulation.2.In order to reduce the need for the number of array elements in DOA estimation and distinguish the incident angle of independent or highly correlated signals,a fast two-dimensional DOA estimation method based on sparse double circular array is proposed in this paper.Firstly,according to the rotation invariance of the two subarrays,the direction of arrival matrix is constructed by using the covariance matrix of the received data of each subarray,and the elevation angle is estimated by using the matrix.Secondly,using the beam transformation theory of circular array,the circular array is transformed into a virtual linear array with Vandermonde structure.Then,the error compensation is carried out for the guidance vector of the sparse array,and a new beam domain guidance vector is obtained by multiplying it with the constructed compensation matrix,which eliminates the error influence caused by the beam space transformation of the sparse array.Finally,the azimuth of the signal is estimated directly by the root MUSIC algorithm.Because the proposed method has no complex spectral peak search,the amount of calculation is small,and the number of array elements used in sparse case is also small.The correctness of the proposed method is verified by computer simulation.