Research On Sparse Imaging Algorithms For Correlated Imaging Systems

It is a brand new radar system of microwave staring correlated imaging system based on the temporal-spatial stochastic radiation field. The new radar system can image the fixed area as well as the traditional radar systems. Also, it obtains high resolution which breaks through the limitation of the actual antenna array aperture, which makes it valuable. Correlated processing is essential for the new radar system as traditional imaging methods are no longer applicable due to the temporal-spatial stochastic behavior of the radiation field. It achieves the target recovery results by combing the received echo signals and the temporal-spatial stochastic radiation field. Several methods for correlated processing are applicable, as an information processing method based on the Gram-Schmidt orthogonalization or regularization. In this paper, correlated processing for sparse targets based on Compressive Sensing (CS) is studied with the sparse prior information about the targets.It is shown in CS theory that signals can be recovered from a few random measured data by the sparse recovery algorithm if the signals are with sparse distribution. In this paper, sparse recovery technology is used for the correlated imaging systems to get better imaging results by applying the sparse prior information about the distribution of scattering points.In this paper, application of sparse recovery algorithms in the correlated imaging systems is studied first. The imaging model of the microwave staring correlated imaging system based on spatial-temporal random radiation field is built, which includes signal generation, radiation, scatter, receiving and correlated processing. Several classical sparse recovery algorithms are used to the correlated imaging radar systems such as Focal Underdetermined System Solver(FOCUSS) method and Sparse Bayesian Learning(SBL) method. A series of simulations are carried out to verify that the system can obtain higher resolution with the use of sparse recovery algorithms.Secondly, the problem of mismatches in radiation field matrix for static targets is studied. Mismatches of locations between scattering points and grids will introduce mismatches in the radiation field matrix, which will lead to the failure of sparse recovery. Given this problem, an imaging model with the presence of mismatches is built. In addition, an improved adaptive sparse recovery algorithm that is based on the constrained least squares(CTLS) technique is proposed. Also, numerical simulations are carried out to verify that the proposed algorithm can simultaneously realize sparse imaging and self-calibration.Finally, imaging of moving targets for microwave staring correlated imaging is studied. A mathematical model for microwave staring correlated imaging of moving targets is established. Extensive research for imaging of moving targets is made. Also, an adaptive sparse recovery algorithm for moving targets that is based on the grid updating is proposed. With an iteration mechanism, the proposed method updates the image and estimates the velocity alternately by sequentially minimizing the Lp norm and the recovery error. Numerical simulations are carried out to demonstrate that the proposed algorithm can retrieve high-resolution image and accurate velocity simultaneously.