| Radar characteristics are the main features of target recognition in the mid-course of the ballistic trajectory.The fretting feature as the unique feature of the target has gradually become a breakthrough in the recognition of mid-course ballistic targets.Based on the establishment of the radar target translation and fretting model,this paper studies the translation parameter estimation and compensation problem,parameter estimation problem with no translation and weak micro-Doppler coupling and parameter estimation problem in sparse sampling with no translation and strong micro-Doppler coupling,respectively.Which provide useful reference and reference for solving the target attack and defense confrontation and target identification in the middle of the ballistic trajectory.The main work and research results of this paper include:(1)The fretting model of the equivalent scattering point of the target in the middle of the trajectory is established.Then,the fretting echo models are established for the pulse system chirp signal and the single-frequency continuous wave signal respectively.(2)Aiming at the problem of high-speed motion compensation of targets in the middle of the trajectory,a translation compensation algorithm based on template matching is proposed.Firstly,512×512 time-frequency map is obtained by binarization and down-sampling.Then convolves the matching template with the time-frequency map to obtain contour-like points.And filters out all true contour points through structural similarity.Lastly,use polynomial interpolation to estimate translation parameters and to realize the translational compensation.In the absence of spectral aliasing,the parameter estimation accuracy and the real-time performance of the proposed algorithm are better than other time-frequency analysis algorithms when the signal-to-noise ratio(SNR)is higher than-4d B.In the case of spectral aliasing,the proposed algorithm has the same applicability,but other time-frequency analysis algorithms fail directly in this case.(3)Aiming at the coupling problem of micro-Doppler curves without translation,a separation algorithm based on curve trend estimation is proposed.Firstly,the algorithm obtains stable and fine binarized curve data through skeleton extraction.Then,the weakly coupled micro-Doppler curves are divided into overlapping regions and non-overlapping regions,and accurate separation of the weakly coupled micro-Doppler curves is achieved by estimating the trend of the micro-Doppler curves in the overlapping regions.Finally,the variational mode decomposition and empirical mode decomposition algorithms are used to decompose each micro-Doppler curve and estimate the corresponding fretting parameters.Simulation experiments show that the algorithm can accurately and quickly separate the micro-Doppler curve when the SNR is greater than-15 d B,which provides support for subsequent parameter estimation.(4)Aiming at the joint estimation of fretting parameters under sparse sampling,a multi-level iterative parameter estimation method based on sparse representation is studied.The traditional method of directly discretizing the dictionary to build a dictionary for sparse solution and parameter estimation,due to the high dimension of the dictionary and the parameters to be estimated are not necessarily located on the discrete grid,the real-time performance is poor,and the accuracy cannot be guaranteed.The multi-level iterative parameter estimation method based on sparse representation separates the fretting angular velocity,reduces the dimension of the dictionary,reduces the discrete interval of parameters through step-by-step iteration,uses the root mean square error to predetermine the fretting angular velocity.The angular velocity estimation value is used to construct a dictionary and sparsely solve it to obtain other fretting parameters.The simulation results show that the time complexity is reduced while improving the accuracy of parameter estimation. |
PDF Full Text Request