FDA-MIMO Radar Target Detection Method With Steering Vector Mismatches

Frequency diverse array-multiple input multiple output(FDA-MIMO)radar comprehensively utilizes the distance dependence of frequency array radar and the waveform of multiple input multiple output(MIMO)radar The diversity feature improves the performance of target detection.However,in the actual target detection process,the steering vector may be mismatched due to factors such as sampling error,waveform distortion,array calibration error,and beam pointing error.This mismatch will seriously affect the detection performance of FDA-MIMO radar.Therefore,this thesis focuses on the problem of FDA-MIMO radar target detection under the condition of steering vector mismatch.The main work is summarized as follows:(1)Aiming at the target detection problem of FDA-MIMO radar under the condition of distance mismatch,a generalized likelihood ratio test(GLRT)detector based on the subspace model is proposed.A subspace signal model is established,and an adaptive subspace detector with or without auxiliary data is derived by using the GLRT criterion,and the target detection under the condition of distance mismatch is realized.The simulation results show that the subspace detector is more robust than the traditional GLRT detector.(2)Aiming at the problem of subspace construction under the condition of distance mismatch,a subspace construction method based on Karhunen-Loève(KL)transformation is proposed.The problem of subspace construction is abstracted as an optimization problem,and the expression of optimal subspace matrix is deduced by using KL transformation,and the adaptive subspace construction under the condition of distance mismatch is realized.Simulation results show that the performance of the subspace detector is related to parameters such as sampling rate,frequency offset,number of transmitting array elements and subspace dimension.(3)Aiming at the target detection problem of FDA-MIMO radar under the condition of distance and angle mismatch,a GLRT detector based on subspace model and a subspace construction method based on KL transform are derived.The simulation results show that the simultaneous mismatch of distance and angle will increase the dimensionality of the subspace where the mismatched steering vectors are located,compared with the case where only the distance mismatch is considered.On this basis,the detection method of high-speed moving target under the condition of distance and angle mismatch is studied.The keystone transform is used to correct the range migration and Doppler spread,the influence of the keystone transform on the noise covariance matrix is analyzed,and a subspace GLRT detector based on the keystone transform is proposed.The simulation results show that when the frequency offset is large,the adaptive detector based on keystone transform has better performance than the traditional detector.