Study On MIMO Radar Waveform Design And Anti-jamming

MIMO radar provides much freedom for radar signal processing,and has gradually developed into a hot research field.With the increasing complexity of battlefield environment,MIMO radar is required to have higher performance.Firstly,in order to prevent the large target from covering the small one,the multi-target detection task requires waveform of the MIMO radar to have lower autocorrelation sidelobe peak level and cross correlation peak level.In addition,complex electromagnetic interference,especially deceptive jamming in the main lobe,causes the deterioration of true target detection performance and the reduction of tracking accuracy,so MIMO radar is required to have higher performance for interference suppression.In order to reduce the sidelobe peak level of waveform and improve the jamming suppression performance of MIMO radar,the research of this thesis is carried out on the basis of existing methods.The main work of this thesis is expressed as follows:In the first part,the method of waveform design for MIMO radar is studied.Firstly,the signal model of phase-coded MIMO radar is established.Secondly,the advantages and disadvantages of classical waveform design methods such as genetic algorithm and simulated annealing algorithm are analyzed.Finally,aiming at the weak ability of heuristic algorithm in local search,a GSQP waveform design algorithm is proposed,which has high performance in local search and low sensitivity to the initial point.The simulation results show that the waveform is optimized with lower autocorrelation sidelobe peak level and cross-correlation peak level by the proposed algorithm.In addition,the influence of weight on autocorrelation sidelobe peak level and cross-correlation peak level is analyzed.The simulation results show that the autocorrelation sidelobe peak level is greatly reduced by controlling weight with a small increase in the cross-correlation peak level.In order to further reduce the sidelobes of pulse compression,the second part focuses on the design of the mismatched filter.Firstly,the signal model of mismatched filter is given,and the performance evaluation of mismatched filter is introduced.The advantages and disadvantages of the existing mismatched filter design methods,such as least squares and convex optimization,are analyzed.Finally,aiming at the limitations of existing mismatched filter design algorithms,a joint design method of waveform and mismatch filter which iscalled ISQP algorithm is proposed.By using convex optimization algorithm,the initial iteration points of SQP algorithm are updated,which reduces the dependence of the algorithm on the initial points and improves the stability of the algorithm.The experimental results show that the proposed algorithm has better performance with less SNR loss.The influence of the code length,waveform number and SNR loss constraint on the algorithm performance is analyzed.In the third part,this thesis mainly studies the application of FDA-MIMO radar in suppression of main-beam deceptive jamming.First of all,the signal model of the deceptive jamming in the main lobe and the FDA-MIMO radar are established,and the distinction between the jamming and the target in the spatial frequency of FDA-MIMO radar is demonstrated.Because the existing methods are sensitive to the errors of target and jamming,a robust anti-jamming method for FDA-MIMO radar is proposed under the premise that range of target is known.Firstly,the proposed method compensates the spatial frequency of target and jamming to ensure that jamming and target are distinguished significantly.Then the target spatial frequency is estimated,and the covariance matrix tapers method is used to update the covariance matrix.Finally,the adaptive beamformer is used to suppress the interference.In the estimation of target spatial frequency,a method to determine the initial range of target is proposed.The simulation results show that the proposed anti-jamming method is robust,and effectively suppress interference even in the presence of errors,which is significantly better than other existing methods.