Research On Sparse Array Calibration And Robust DOA Estimation Algorithm

Direction of arrival estimation has always been an important part of the field of array signal processing,and has important engineering application value in many military and civilian fields such as wireless communication,radar,unmanned driving,and ocean exploration.In recent decades,sparse arrays have become the mainstream research direction by virtue of the advantages of obtaining a larger array aperture with a smaller number of sensors,realizing underdetermined direction of arrival estimation and improving estimation performance.From the perspective of practical applications,this thesis focuses on Array calibration and robust high-precision angle estimation algorithm when sparse arrays(especially nested arrays,extended coprime arrays)have model errors,and mainly study the mutual coupling effect modeling between nested array elements and the countermeasures against mutual coupling.The robust DOA estimation algorithm based on the non-uniform sparse array,the mixed error model modeling of the amplitude and phase error and the position error of the array element in the nonuniform sparse array,and the high-precision direction finding algorithm that can resist the mixed error,the bistatic coprime array MIMO radar has the channel amplitude and phase error.Array calibration algorithm and joint estimation of launch angle and angle of arrival.The research content is summarized as follows:(1)For nested arrays,an algorithm that can resist the error caused by mutual coupling and achieve robust DOA estimation is proposed.According to the characteristics of the array manifold matrix of the nested array,the algorithm gives the matrix transformation of the product of the mutual coupling matrix and the steering vector of the nested array,and separates the mutual coupling vector from the new array manifold which is not affected by the mutual coupling coefficient.The simplified block sparse recovery model constructed by the new array manifold can effectively eliminate the error caused by the mutual coupling between the array elements,improve the DOA estimation accuracy,and use the difference array of the nested array to improve the degree of freedom and realize the estimate.Simulation results corroborate that the proposed algorithm can estimate the angle when the number of sources exceeds the number of array elements.At the same time,compared with other existing DOA estimation algorithms that are insensitive to mutual coupling,the proposed algorithm has higher estimation accuracy.(2)For a sparse array,a mixed error model with both amplitude and phase errors and array jitter is constructed.Taking the extended coprime array as an example,according to the diagonal characteristics of its amplitude and phase error and array position error matrix,the mixed error is equivalent to the matrix deformation.It is modeled as the amplitude and phase errors related to the incident angle,and then a new array flow pattern that is not affected by the error is constructed through the matrix structure of the new error model.Finally,the DOA estimation is realized by using the spatial sparsity building block sparse recovery problem of the received signal,the algorithm is insensitive to array mixing error and has more robust estimation results.The numerical experimental results show that the proposed algorithm has more advantages than the existing algorithms in the number of angle estimates and the accuracy of estimation results.(3)An array calibration algorithm of bistatic coprime MIMO radar is proposed to realize a high-precision DOD-DOA joint estimation algorithm.Considering that both the transceiver arrays of bistatic coprime MIMO radars have amplitude and phase errors,a high-power calibration source is used to separately calibrate the algorithm.The transmit and receive arrays are calibrated for errors.After calibration,the augmented receive and transmit correlation matrices are constructed respectively.The DOD and DOA are estimated by two one-dimensional MUSICs,respectively.Finally,the estimated DOD and DOA are paired by spatial filtering.Simulation experiments verify the effectiveness of the algorithm through spatial spectrum,root mean square error curve and angle matching results.