Study On Forward-looking Super-resolution Imaging Technology Of Small Array Radar

In many civil and military fields,such as precision guidance,autonomous landing,terrain mapping,etc.,radar forward-looking imaging is very important.The traditional forwardlooking imaging method based on real-beam scanning makes it difficult to obtain highresolution images due to the restriction of the actual radar aperture.Compared with the whole imaging scene,the target of interest usually occupies only a small part of the area.This sparsity makes compressed sensing(CS)applicable to high-resolution,forward-looking image reconstruction.However,the strong noise and mesh mismatch in the radar echo affect the quality of the image generated by the CS method.This paper focuses on the forwardlooking imaging technology of unmanned aerial vehicles(UAVs),missiles,and other small moving platform radars and studies the CS-based super-resolution imaging methods of the single-input single-output(SISO)system and the multiple-input multiple-output(MIMO)array system radars,aiming at improving the forward-looking imaging resolution,noise robustness,and sparse reconstruction accuracy.Specific research contents include the following aspects:1.Aiming at the noise robustness of the CS-based super-resolution imaging method in the SISO radar system,a forward-looking imaging method combining echo denoising and noise weighting is proposed from the perspective of echo denoising preprocessing.In the proposed method,the complex variational mode decomposition(CVMD)is used to preprocess the original radar echo signal to improve the signal-to-noise ratio(SNR),and then the weight generated by the CVMD denoising process is used to weight the norm to optimize the contribution of noise and signal to the CS optimization function.The robustness of forwardlooking super-resolution imaging is improved by two steps: echo denoising and noise weighting.Finally,simulation and measured data are used in the experiment.Compared with the existing super-resolution imaging methods,the effectiveness of our proposed method in improving the resolution of forward-looking imaging and having stronger noise suppression ability is verified.2.Aiming at the noise robustness of the CS-based super-resolution imaging method in the SISO radar system,a forward-looking imaging method combining sparsity and echo low rank(CSELR)is proposed from the perspective of using more prior information.In CSELR,in addition to the sparsity of the target commonly used in the traditional CS-based imaging algorithm,the inherent relationship between the echo signals is used to improve the image quality.The low rank of the echo refers to the energy of the echo signal,which can be represented by several large eigenvalues when noise is not considered.In addition,an algorithm based on Augmented Lagrange Method(ALM)and Alternating Direction Method of Multipliers(ADMM)is proposed to effectively solve the dual optimization problem in the proposed method.Finally,based on the simulation and measured data,we prove that the proposed CSELR method can not only achieve forward-looking super-resolution imaging but also has strong noise robustness.3.Aiming at the noise robustness of the CS-based super-resolution imaging method in the SISO radar system,a forward-looking imaging method combining sparsity and image lowrank(CSILR)is proposed from the perspective of using more prior information.In the CSILR method,not only the sparsity of the target is considered but also the low-rank characteristic of the final generated image.The low-rank image means that there are all 0rows or all 0 columns in the high-quality forward-looking image,and the false strong scattering points can be suppressed by the low-rank constraint on the image.Then,the ALM method is used to solve the double constraint problem in the proposed model under the ADMM framework,and a robust forward-looking super-resolution image can be obtained.Finally,the experimental results show that compared with the current CS-based imaging method,it can not only break through the limitation of real beam resolution to achieve superresolution imaging but also has strong noise robustness,thus ensuring the reconstruction performance of the target.4.Aiming at the problem of forward-looking super-resolution imaging in the MIMO radar system,the orthogonal waveform design and array structure design of the MIMO radar are analyzed first,which are the basis for building the MIMO radar imaging system.Then,aiming at the problem of the poor robustness of the MIMO radar sparse imaging in the strong noise environment,a forward-looking imaging method of the MIMO radar combining sparse and double low-rank(CSDLR)is proposed.One of the low-rank characteristics refers to the energy of the echo signal.When noise is not considered,it can be represented by several large eigenvalues,i.e.,the echo signal has low-rank characteristics;another low rank refers to the inherent characteristics of the forward-looking radar image.Finally,the effectiveness and robustness of the proposed method are verified by simulation and real-data experiments.In fact,compared with SISO radar,MIMO radar has more degrees of freedom.More degrees of freedom means that the collected samples can provide more information,so MIMO radar has more advantages in realizing forward-looking super-resolution imaging.5.Aiming at the problem that the reconstruction accuracy of the CS-based forward-looking imaging method is poor when the mesh is mismatched,a meshless sparse reconstruction forward looking imaging method based on atomic norm minimization(ANM)is proposed from the perspective of continuous parameter estimation.The proposed method completely abandons grid division,adopts the meshless signal reconstruction method based on atomic norm minimization,and uses convex optimization theory to transform the atomic norm minimization problem into a semi-positive definite programming problem.Finally,Vandermonde used the obtained covariance matrix to obtain the estimated frequency and amplitude of the target,where the frequency corresponds to the target position information.In the experiment part,combined with the echo denoising method of CVMD proposed in Chapter 3,we propose a joint echo denoising and atomic norm minimization method,which improves the robustness of imaging under strong noise.