The Information Processing Methods And Simulations In Microwave Staring Correlated Imaging

In Microwave Staring Correlated Imaging, radar is located in a stationary platform and imaging the fixed area. It can realize continuous gaze and imaging of a specific area. The angle resolution of the traditional staring imaging, ie. the real-aperture imaging is limited by the actual antenna array aperture. Therefore, the research of an imaging system which can realize staring imaging, meanwhile break through the limitation of the actual antenna array aperture, is of great significance.A brand new microwave staring correlated imaging system based on the temporal-spatial stochastic radiation field is studied in this dissertation. Under the temporal-spatial stochastic radiation field, we can obtain more effective observation samples of the target information. Combining the received echo signal and the temporal-spatial stochastic radiation field, the distinguishment of the targets within the beam coverage is achieved. In the first section, the mathematical model of microwave staring correlated imaging is established by the analysis of the physical process and the method of time-domain electromagnetic derivation. Next, the difference of the traditional radiation field and the temporal-spatial stochastic radiation field is discussed. To characterize the temporal-spatial stochastic behavior of the radiation field, the effective number of the radiation field’s eigenvalues is first proposed from the points of the degree of freedom of the radiation field and the optimal resolution, respectively. At last, according to the above theoretical analysis, a novel approach to constructing the temporal-spatial stochastic radiation field by the random frequency-hopping transmitting signal is proposed.Due to the temporal-spatial stochastic behavior of the radiation field, the direct imaging method is not suitable here. Correlated processing is essential to achieve the target recovery results by combing the received echo signal and the temporal-spatial stochastic radiation field. Moreover, the complexity of the transceiver configuration limits the application of the conventional imaging method to correlated processing. In the second section, the information processing is refered as inverse scattering problem and a novel information processing method based on the Gram-Schmidt orthogonalization is proposed. In order to solve the ill-posedness of the target recovery under the noisy situation, regularization is introduced here, including truncated singular value decomposition(TSVD), Tikhonov regularization, total variation(TV) regularization. Finally, the effectiveness of the proposed imaging algorithms is validated through the numerical simulations.Information processing for sparse targets based on Compressive Sensing(CS) is studied in the third section. Using the sparse prior information about the targets, higher spatial resolution can be achieved using less radiation field samples. The mathematical models of microwave staring correlated imaging and CS are compared and the Off-Grid problem is emphasized when existing CS reconstruction algorithms are applied to the information processing of microwave staring correlated imaging. Good sparse recovery performance is based on the assumption that the scatterers are positioned at the pre-discretized grid locations; otherwise, the performance would significantly degrade. Two new sparse recovery algorithms are proposed for the general Off-Grid microwave staring correlated imaging from the aspects of the greedy iteration and Bayesian statistical optimization, respectively. The first is, least-square recovery method based on iterative Lo Norm minimization (iLo-LS), and the second is, sparse adaptive calibration recovery via iterative maximum a posteriori (SACR-iMAP). Numerical simulations are carried out to verify the effectiveness of the proposed method.The above sections are all aimed to static target recovery in the staring imaging area. The fourth section will study moving target information processing methods in microwave staring correlated imaging. First the microwave staring correlated imaging model according to the moving targets is established. The influence of the target motion to the foregoing correlated processing methods performance is analyzed. Then, we propose two information processing methods aiming to the moving targets, which is named, the imaging algorithm based on updated overcomplete dictionary approach and adaptive sparse recovery algorithm based on velocity estimation. In the methods proposed, the scatter motion and the reflectivity are estimated in an optimal and global way. Simulation results verify the validity of the proposed imaging methods.