Research On 3D Imaging Method Of Tomographic SAR Structures Based On Spectral Estimation

Synthetic Aperture Radar(SAR)is a remote sensing imaging method.Due to its outstanding advantages of all-weather,all-weather,and high imaging accuracy,it has been applied in many practical fields such as geological monitoring,urban development,military strikes,and occupies an important position in the field of remote sensing.Synthetic aperture tomography(Tomo SAR)is a multi-baseline interferometry technology that can increase the number of baselines on the basis of traditional two-dimensional satellite images,forming a synthetic aperture in the altitude direction,completing altitude tomography,estimating the power spectrum pattern(PSP),or backscatter coefficient,to achieve three-dimensional inversion imaging,Due to its high imaging accuracy and low computational complexity,this technology is currently the focus of research in the field of synthetic aperture radar;At the same time,the demand for obtaining massive high-resolution SAR images has led to the development of tomographic SAR technology.The conditions for conducting research on three-dimensional imaging of tomographic SAR structures are ripe,making it necessary to develop research on tomographic SAR technology in China.This paper focuses on the research of tomographic SAR algorithm based on signal processing and its adaptation in actual imaging scenes.Firstly,by consulting a large number of relevant literature,this paper summarizes the research background,current situation,and application of three-dimensional tomographic SAR imaging technology in urban structures.Through comparative analysis,it is understood that the current problems in the field of tomographic SAR imaging still exist in terms of technology and algorithms,including insufficient number of passes,large pass intervals,and uneven baseline distribution.It is pointed out that the current research direction should be high-precision imaging under the above non ideal conditions.Secondly,through theoretical analysis of the imaging principle,processing flow,and imaging resolution of tomographic SAR three-dimensional imaging technology,combined with simulation experiments under different navigation parameters,the influencing factors of tomographic SAR imaging are clarified,providing theoretical support for subsequent algorithm research and improvement.Then a brief introduction is given to the three algorithms currently mainly used in tomographic SAR three-dimensional imaging and their imaging processes.Among them,the compression sensing algorithm reconstructs sparse and scattered objects,but is limited by the number of scatterers;The beamforming algorithm based on signal processing does not consider the number of scatterers,but directly determines whether the scatterers are targets or noise based on their power spectrum strength.Focusing on a class of algorithms based on beamforming theory,including Capon and DCRCB algorithms,from the perspective of noise suppression,through weighting processing,the constraint of minimizing the noise variance of the signal output is calculated to obtain the optimal weighting,thereby achieving the estimation of backscattered signals,which can improve the shortcomings of Fourier transform imaging such as low resolution,high sidelobe,and large impact of imaging effects by system errors,and has super resolution capabilities.At the same time,simulation experiments are conducted to verify the imaging accuracy and robustness of the above algorithms in various imaging scenarios.Finally,on the basis of previous research,in order to further improve the imaging accuracy in actual imaging environments,this paper proposes an improved beamforming optimization algorithm,which combines the L1 norm constraint function with the Double Constrained Robust Capon Beamforming algorithm(DCRCB),The cost function of the Alternating Direction Method of Multipliers(ADMM)is constructed to further sparsely optimize the DCRCB recovered backscatter coefficients and achieve three-dimensional imaging of tomographic SAR.The ADMM algorithm is based on the augmented Lagrange algorithm,which transforms a more complex global solving problem into two or more simple local subproblems that are easier to solve.In the iteration of the ADMM algorithm,each subproblem can perform sparse reconstruction and noise reduction operations respectively.The algebraic expressions of the separated local subproblems are relatively simple,and they can easily find a determined solution without requiring convergence and constraint operations.Therefore,the ADMM algorithm has the advantage of high reconstruction accuracy.In this paper,the 8-channel airborne array interferometric SAR data in Yuncheng,Shanxi Province released by the Institute of Aerospace Information Innovation,Chinese Academy of Sciences in 2021 are used for experimental verification.The experimental results verify the effectiveness of the algorithm.