Analysis And Estimation Of Uncertainty Parameters Of The Transceiver Channels In Microwave Staring Correlation Imaging System And Research On Imaging Methods

Microwave staring correlated imaging(MSCI)is a novel high resolution radar imaging technology in staring imaging geometry by utilizing two-dimensional temporal-spatial stochastic radiation field(TSSRF).MSCI employs random radiation source(RRS)to construct TSSRF at the imaging area,and the target images are recon-structed by correlation process(CP)of the echo signals and the computed TSSRF.As a computational imaging technology,MSCI needs to accurately obtain the parameters of the radar system at the imaging working state,so as to accurately compute the TSSRF matrix.The final image of the target is obtained through correlation process.However,the uncertainty of the radar system parameters greatly affects the imaging quality of MCSI,which is also an obstacle for MSCI from theoretical research to practical appli-cations.This paper mainly studies the uncertain parameters in the transceiver channels of the MSCI system.The first part of this paper focuses on the radar systems consisting of multiple trans-mitters and single receiver which transmit random frequency-hopping signals.Then the uncertain parameters in the transceiver channels of the MSCI system are elaborately an-alyzed,and the imaging model containing the uncertain parameters is established.The influence of uncertainty parameters on MSCI is analyzed.And the influence under dif-ferent error ranges of different uncertain parameters on imaging is analyzed through simulations,so as to obtain the calibration accuracy of the MSCI system.The second part of this paper studies the high-robust MSCI methods.In order to reduce the illness of the solutions for MSCI,we consider introducing sparse priors for the sparse targets and adding appropriate regularization constraints for the non-sparse targets.For sparse targets,this paper considers the spatial cluster characteristics of the scattering points,and proposes a new spatial cluster distribution hypothesis prior.The iterative process is derived using Variational Bayesian Inference(VBI)and the Gener-alized Approximate Message Passing(GAMP)framework,and the effectiveness of the proposed algorithms is verified through simulations.For non-sparse targets,a nonlo-cal low rank regularization term based on Gaussian mixture classification is proposed.By classifying the nonlocal similar sub-images of the imaging scene and adding low-rank constraints,the quality of the images is improved.The simulation verifies that the proposed methods have higher tolerance to uncertain parameters and the imaging performance is more robust.The third part of this paper studies the uncertainty parameter estimation methods based on the auxiliary calibration channel.First,the uncertainty parameter calibartion system design scheme based on the auxiliary channel is proposed.Then the specific data processing procedures and data processing methods are proposed while employing the parametric model of the group-delay paramters,and simulations verify the effec-tiveness of the proposed methods.Finally,the experiments in the anechoic chamber preliminarily verify that the proposed methods can effectively estimate the uncertainty parameters,thereby reducing the calculation errors of the TSSRF.The fourth part of this paper studies the self-calibration methods of MSCI using iterative interal search.The self-calibration methods for baseband time synchronization errors and random phase errors are considered seperately.In order to reduce the com-putational complexity and achieve a better convergence result,an iterative estimation method for uncertainty parameters based on batch processing is utilized.The effective-ness of the proposed methods is verified by simulations,which further complements and perfects the processing methods for the uncertainty parameters of MSCI.