Research On Ionospheric Clutter Mitigation Method For Hfswr Based On Sparse Polarization Array

The high-frequency surface wave radar(HFSWR)works in the short-wave band(3 ? 30 MHz),and can achieve over-the-horizon detection of sea-surface targets and low-altitude aircrafts.Therefore,it is a good choice of the real-time and large-coverage surveillance for the exclusive economic zone(200 sea-mile)of a nation.However,the electromagnetic environment for the HFSWR is quite complex,and the HFSWR is susceptible to a variety of external interferences such as short-wave radio,sea clutter,ionospheric clutter and industrial interference,which limit the target detection performance of HFSWR.Among these external interferences,ionospheric clutter has become a major factor affecting the detection performance of HFSWR due to its long existence and wide coverage.Existing methods for clutter mitigation are mainly based on the characteristics of the ionospheric clutter in the time-frequency domain and the space-polarization domain,and have achieved some good results.However,due to the constraints of the performance of the practical received array and array nonideality(e.g.,phase and amplitude error,mutual coupling,etc.),the suppression effects of these methods are limited.This thesis focuses on the ionospheric clutter in the HFSWR.From the perspective of the characteristics of the polarization received array,it is committed to research antijamming algorithms that conform to the received data from actual array,and study sparse array design with larger aperture which does not increase the hardware cost and complexity of the received system.On this basis,further research on ionospheric clutter suppression method is conducted with the polarization sparse array model.The main research contents of this thesis are as follows:Firstly,this thesis studies the anti-jamming method based on a collocated dualpolarized L-array.The collocated polarization array consists of mutually orthogonal uniform linear arrays,in which a collected dual polarization vector sensor is placed on each array element.In order to make full use and excavation of all the information of the ionospheric clutter,this thesis proposes to establish a multi-domain subspace which extracts the difference between the clutter and the target on each parameter,and then to construct a multi-domain collaborative filter based on the oblique projection operator forclutter suppression.Even if the difference of each parameter between the clutter and the target is very small,the proposed filter can obtain a good performance of clutter suppression.At the same time,this thesis analyzes the performance and error of the proposed filter under a collocated polarization array.In particular,for the case where the amplitude of the actual array is inconsistent,an improved multi-domain collaborative filter based on the polarization domain is proposed to obtain better clutter suppression performance.Secondly,without increasing the hardware cost of the radar system,this thesis introduces sparse scalar-sensor linear arrays into the field of polarization sensitive arrays,and proposes spatially separated sparse polarization linear arrays to obtain longer apertures for improving parameter estimation performance and filtering performance.The sparsely polarization linear array proposed in this thesis is not directly extended from the sparse scalar-sensor linear array,but arranges different types of sparse scalar-sensor arrays according to a certain rule to enlarge the aperture as much as possible.At the same time,the proposed array meets the requirements of received array of the HFSWR,and reduces adverse effects of two mutual coupling(i.e.,the mutual coupling between array elements and the mutual coupling between differently polarized antennas in a collected vector sensor).This thesis discusses the effects of completely and partially polarized signal models on the array design.Some difference co-arrays of the proposed spatially separated dual-polarized arrays(i.e.,virtual arrays)are non-uniform linear arrays,which will degrade the performance of parameter estimation and filtering.To solve this problem,a matrix reconstruction method based on the oblique projection operator is proposed to fill all the holes in difference co-arrays,and then the virtual covariance matrix with increased degree of freedoms(DOF)can be achieved for improvement of the parameter estimation performance and filtering performance.In particular,the matrix interpolation based method is proposed to mitigate the adverse effect of unknown nonuniform noise.Then,in order to reduce the effects of two mutual couplings,this thesis proposes a spatially separated tri-polarized array with lower mutual coupling.The design idea of the array is to first give a sparse scalar-sensor array design to reduce the mutual coupling between array elements,and then use the sparsity of such array to spatially separate orthogonally polarized antennas according to a certain rule,for reducing the mutual coupling betweenorthogonally polarized antennas.Moreover,the proposed array design provides different schemes for different scenarios with received array,which increases the flexibility and application of the array design.Finally,the sparse linear array is applied to the two-dimensional array field in this thesis,e.g.,L-shaped array composed of two orthogonal sparse subarrays.According to the property of the sparse L-shaped array,this thesis presents an automatic angle-matching and matrix reconstruction based method to obtain a virtual covariance matrix with increased DOF,for enhancing the performance of two-dimensional DOA estimation and filtering.Therefore,this thesis combines the sparse L-shaped array with the polarization sensitive array,to construct a spatially separated polarization L-array in practice,and then to propose a two-step method with parameter estimation and collaborative filtering for ionospheric clutter suppression.For the parameter estimation,the vertically and horizontally polarized antennas in the actual array are placed in different array elements to reduce mutual coupling.The proposed algorithm first estimates the two-dimensional DOAs,and then uses the estimated DOAs to compensate the spatial information,introduced from spacing between orthogonally polarized antennas,in the estimation of polarization phase delay.In collaborative filtering,the filter construction model is matched with the received data from actual array,and a polarization-domain adaptive multi-domain collaborative filter is constructed to eliminate the phase inconsistency among vertically polarized antennas in the zenith direction,for achieving good performance of clutter suppression.The measured data verified the effectiveness of the proposed two-step method.By effectively suppressing the ionospheric clutter,the detection performance of the HFSWR is further improved.