Research On ISAR Fusion High-resolution Imaging Technology For Space Target

In recent years,the number of space targets has increased rapidly with the development of space technology,the space environment is becoming more and more complex,and the effective monitoring,tracking and recognition of space targets have great military and economic benefits.The multi-radar signals fusion imaging is an effective method to improve radar slant-and cross-range resolution,by which the finer structure features of space targets can be obtained and the performances of space targets recognition can be improved.Aiming at the key theory and technical difficulties in signal fusion for high-resolution ISAR imaging of space target,this dissertation makes research on earth oriented three-axis-stabilized space target echo simulation,ISAR imaging processing using precise ephemeris,sparse subband fusion for high slant-range resolution,aperture fusion algorithm for high cross-range resolution,optimized location of multi-station and multi-angle fusion imaging,which provide theoretical support and practical basis for space target ISAR fusion imaging.First,the ISAR baseband echo model of space target based on two body model is constructed under consideration of earth oriented attitude stabilization and high-speed motion,which is used to analyze accurately the motion characteristics of space target and the influences on ISAR imaging.According to the baseband echo model,the effect of the high-speed motion of the target on the matched filtering is analyzed,and the motion compensation is realized by using the speed parameters extracted from the precise ephemeris.Furthermore,combined with the characteristics of the target's orbit,monostatic ISAR imaging plane vectors are derived and the range history of scattering center is described precisely based on analyzing the space-variant property of ISAR imaging plane,a correction method of the migration through two dimensional resolution cell for space target in nonideal imaging section by using precise ephemeris is proposed,which improves the image focusing degree.Secondly,the sparse subband fusion model for linear frequency modulation(LFM)echo signals with the matched filtering and dechirp processing is established.The nonparametric fusion method for uniform gapped sampling and frequency domain reconstruction are discussed,then a sparse nonuniform fusion and uniform reconstruction in frequency domain based on GSPICE method are proposed,which combines the l₁ norm constraint and the cyclic iterative optimization,avoids the uniform interpolation processing,and improves the precision and convergence speed of the frequency domain reconstruction.Further,for the noncoherence problem,the mutual coherence processing based on GSPICE method and subband fusion for two dimensional image processing frame are proposed,which combines time domain coherence and parameter search,considers both the translational motion compensation and the azimuth focusing,and ensures the accuracy and efficiency of the incoherent sparse subband fusion.Thirdly,for the three-axis-stabilized space target co-site sparse apertures fusion problem,two optimal solution models under l₁ norm and weighted l₁ norm constraints are derived based on sparse aperture fusion imaging model,then the complex Bayesian compressed sensing(BCS)reconstruction with the Gaussian priors considering the residual space-variant phase error is analyzed,further,the complex Bayesian Compressive sensing autofocus algorithm using Laplace priors is proposed for co-site sparse apertures fusion.Compared with the weighted l₁ norm constraint reconstruction and the Bayesian Compressed Sensing algorithm with the Gaussian priors,the proposed algorithm has stronger sparsity enhancement and better fusion imaging quality.Finally,aiming at the multi-angles fusion of different sited radars for high resolution imaging problem,the influence of relative spatial-variation between imaging planes on fusion imaging is analyzed,a new method of optimized location based on precise ephemeris is proposed.Then multi-station echoes fusion imaging is realized by using BP algorithm and the correlation between fusion station optimized location and cross-range resolution improvement is verified.Furthermore,the noncoherent problem of echo sets from different radars is analyzed,and the migration through two dimensional resolution cells is corrected by using cooperative scaling factor,combined with Bayesian compression sensing self-focusing algorithm,the phase error compensation and fusion imaging are realized,and the quality of ISAR image is improved.