Study On Joint Target Parameter Estimation Methods In Polarimetric Bistatic MIMO Radars

As a new system of radar, MIMO radar has the tremendous advantages in manyaspects, such as target detection and localization. Therefore, many researchers pay a lot ofattention to it. Target parameter estimation is an important content in MIMO radar signalprocessing, and it is also a key technology in target localization and target imaging, so ithas broad application prospects.Conventional bistatic MIMO radar can improve the resolution of DOD and DOA byforming big virtual aperture. However, when targets are closely spaced in spatial domain,the estimation performance is deteriorated. Thus, polarization-sensitive array is used asthe receive array, not only the angle information but also polarization information oftargets can be obtained. The polarization characteristics are observable physicalcharacteristics of the electromagnetic wave, we can achieve high-resolution estimation oftargets by making fully use of polarization information and the estimation performancewill be improved. On the basis of joint estimation methods for DOD and DOA intraditional biatatic MIMO radar, the paper will focus on joint target parametersestimations methods in polarimetric bistatic MIMO radar. The work is supported by theproject “Joint parameter estimation and target localization based on polarization diversityin MIMO radar（”No.61071140）, which is a National Natural Science Foundation project.The main innovation of the research is as follows:A bistatic MIMO radar signal model based on polarized linear receiver array isdescribed. An algorithm for joint estimation of angle, polarization parameters andDoppler frequency in bistatic MIMO radar based on propagator method is proposed. ThePM avoids the singular value decomposition（SVD）of the covariance matrix of thereceived signals, the computational complexity is reduced, so the PM is a fast algorithm.The simulation result shows that the proposed method has a high-resolution by usingpolarization information fully, especially when targets are spaced too closely in spatialdomain to distinguish them.Due to the defect that traditional bistatic MIMO radar only can estimate onedimensional transmit and receive angles（1D-DOD and1D-DOA）, we set up a model byintroducing uniform planar array as transmitter array and receiver array in the polarimetric bistatic MIMO radar. An ESPRIT-based algorithm for joint estimation of2D-DOD、2D-DOA and polarization parameters in bistatic MIMO radar is proposed andthe algorithm does not need multiple-dimensional spectrum peak searching. Thesimulation results verify the proposed method has high-resolution by using polarizationinformation fully, it can estimate each parameter of targets, which are spaced closely inspatial domain, so it can achieve three-dimensional positioningIn order to avoid additional parameter pairing of ESPRIT algorithm and reduceconstructed covariance matrix and the eigenvalue decomposition of the cumulative error,on the basis of polarized square array model in bistatic MIMO radar, we proposed a newalgorithm for joint estimation of2D-DOD、2D-DOA, polarization parameters andDoppler frequency in bistatic MIMO radar based on parallel factor （PARAFAC）method. The proposed method can effectively identify multiple parameters of the targetsand the parameters can be paired automatically. Owe to the use of polarization diversity,it can accurately estimate targets which are spaced closely in spatial domain, especiallywhen multiple targets have similar2D-DOAs, similar2D-DODs, even the same2D-DOA.