Research On Detection And Parameter Estimation For High-speed And High-maneuvering Radar Targets

The emergence of high-performance fighter,cruise missile and hypersonic aircraft has made the detection of high-speed and high-maneuvering targets one of the important tasks of radar and of great significance in modern warfare.In order to improve the reliability of target detection results and the accuracy of parameter estimation,the coherent integration is usually adopted to increase the signal-to-noise ratio(SNR)of the echo.However,the complex motions of high-speed and high-maneuvering targets result in the range migration and Doppler frequency migration of the echo,which will affect the coherent integration gain seriously and make the traditional coherent integration methods which are developed for constant speed targets invalid.Therefore,the detection and parameter estimation algorithms based on coherent integration which are suitable for high-speed and high-maneuvering targets are studied in this paper.The main research contents and achievements of this paper are as follows:(1)The detection and parameter estimation algorithm for high-speed and high-maneuvering targets based on iterative adjacent cross correlation function(ACCF)is studied.Firstly,the iterative ACCF is exploited to correct the range migration and Doppler frequency migration.Then,by using the fast Fourier transform(FFT),the coherent integration and parameter estimation can be easily achieved.The shortcoming of parameter estimation is that the FFT requires a large number of zero-padding.To overcome this shortcoming,the improved iterative ACCF method based on Chirp-Z transform is proposed.The proposed method can improve the parameter estimation precision with much lower complexity.The simulation and analysis results show that the coherent integration method based on iterative ACCF has low computational complexity but poor anti-noise performance.(2)The detection and parameter estimation algorithm for high-speed and high-maneuvering targets based on the generalized Radon-Fourier transform(GRFT)is studied.GRFT can obtain the ideal coherent integration results,but it has blind speed side lobe(BSSL)phenomenon and large computational burden.By converting the realization of GRFT into an optimization problem in parameter space,this paper proposes the BSSL learning-based modified wind driven optimization algorithm to fast implement GRFT.The simulation results indicate that the proposed method greatly reduces the computational complexity of GRFT.Compared with the previously proposed GRFT fast algorithm based on particle swarm optimization,the proposed method is able to significantly improve the anti-noise performance with a slightly larger complexity,which is more efficient.(3)The detection and parameter estimation algorithm for high-speed and high-maneuvering targets based on the time reversing transform(TRT)is studied.The coherent integration algorithm based on TRT and non-uniform fast Fourier transform is proposed in this paper and the cross term suppression ability and the computational complexity of the propsed method are also analyzed in detail.Compared with another TRT-based method,the proposed algorithm can obtain slightly better anti-noise performance with lower computational complexity.Compared with the algorithms in the previous two chapters,the proposed method can achieve good balance between anti-noise performance and computational complexity,and has excellent multi-target resolution.In summary,the three algorithms studied in this paper have their own advantages and disadvantages,and thus have different application scenarios.The first algorithm has the lowest complexity but the worst anti-noise performance,which is suitable for mono-target scene with high SNR to quickly estimate the motion parameters of the maneuvering target.The second algorithm has the strongest anti-noise performance and is suitable for weak target detection scene,but the performance of it will suffer some loss in multi-target scene.The computational complexity and anti-noise performance of the third algorithm are both between the first two algorithms.Due to the excellent multi-target resolution and avoidance of BSSL,the third algorithm is suitable for multi-target scene.