Method Of High Resolution ISAR Imaging And Scaling For Maneuvering Target

Inverse Synthetic Aperture Radar(ISAR)has been widely used in military and civil fields because ISAR can observe the target from long range under all-time and all weather conditions.However,due to the inhomogeneous translation and rotation motions of maneuvering targets,traditional ISAR imaging methods designed for stationary targets face the following problems: Firstly,the inhomogeneous translation motion of maneuvering targets reduce the performance of existing parameterized translational motion compensation methods.Especially under the low SNR case,we cannot obtain focused ISAR images of maneuvering targets.Secondly,inhomogeneous rotation motion of maneuvering targets will induce the high order phase and the migration through range cells(MTRC),resuting in blurring ISAR images.Finally,the inhomogeneous rotation motion of maneuvering targets make it difficult to estimate the rotation motion parameters and obtain accurate scaled images.To tackle these problems,this thesis proposes a particle swarm optimization(PSO)-based translational motion estimation and compensation method,which facilitate obtaining high resolution images of maneuvering targets.Then,by utilizing the PSO and Genetic Algorithm(GA)algorithms respectively,two novel non-uniform rotation motion estimation and compensation methods are proposed.After that,the accurate scaled ISAR images can be obtained.The major contents are arranged as follows:1.The basic principle of ISAR imaging is introduced.Common methods of translational motion compensation and MTRC correction based on the first order Keystone transform are presented.Then maneuvering targets' movement model and ISAR imaging signal model are constituted.Through the simulation results,the problems faced by the stationary target ISAR imaging methods are clarified when processing maneuvering targets ISAR imaging.2.To achieve the accurate estimation and compensation of the translational motion of maneuvering targets under long integration time,we propose a PSO-based method to estimate and compensate the translational motion.Firstly,the long-term ISAR echoes are divides into short apertures,in which the translational motion of the maneuvering target is analyzed and modeled as a second order polynomial.Secondly,taking the image entropy as the cost function,the parameters estimation of the translational motion is modeled as an unconstrained optimization problem.PSO is utilize to solve the optimization problem and estimate the translational motion parameters.Then,the translational motion can be compensated and the high resolution image of the corresponding short aperture can be obtained by orthogonal matching pursuits(OMP).Finally,the effectiveness and robustness of the proposed method are verified by the experiments using simulated and measured data.3.To deal with the compensation of high order phase terms and MTRC induced by non-uniform rotation motion of the maneuvering target,we propose the non-uniform rotation motion estimation and compensation methods based on PSO and GA,respectively.The experiments have shown that focused ISAR images can be obtained by performing proposed algorithms on simulated and electromagnetic data.Firstly,the constant acceleration model of the non-uniform rotation motion of the maneuvering target is constructed.By taking the image entropy as the cost function,the parameter estimation of uniform rotation motion can also be modeled as an unconstrained optimization problem.Since PSO and GA algorithms have simple structures and high efficiency,they are utilized to estimate the rotation motion parameters.By using the resampling technique,the non-uniform rotation motion can be transformed into uniform rotation motion.The remaining high order phase term is compensated by constructed the corresponding compensation term.After the above processing,the coupling of range and cross-range dimensions has been removed.The first order Keystone transform can be applied to eliminate the MTRC and focused ISAR image can be obtained by the cross-range compression.Finally,the accurate scaled ISAR image can be realized based on estimated roationa motion parameters.According to the experimental results of simulated and electromagnetic data,the validity and robustness of the proposed methods are verfied.