| Synthetic aperture radar ground moving target indication(SAR-GMTI)technology is an important research direction in the field of SAR,which can achieve dynamic object detection,tracking and positioning at the same time,but also obtain highresolution images of moving targets and scenes,with a wide range of application prospects.Traditional SAR-GMTI technology usually requires the SAR platform to fly in a straight line at a uniform speed in order to obtain a more satisfactory processing result.However,when the SAR platform maneuvers(such as acceleration/deceleration,turning,pulling up,dive),the 3-dimensional acceleration of the platform will cause uneven sampling of azimuth airspace,channel equivalent baseline time change,Doppler modulation frequency change and other issues,and the use of traditional SAR-GMTI method processing will seriously affect the detection effect,increase the parameter estimation error,and cause motion target image distortion.Therefore,it is necessary and significant to study the SAR-GMTI method suitable for mobile trajectory platforms.At present,the research in the field of maneuver trajectory SAR-GMTI is still in the preliminary stage,and the proposed method is mainly only for a specific maneuver trajectory,such as the azimuth constant acceleration maneuver trajectory.Therefore,this thesis focuses on the establishment of a SAR-GMTI unified signal model suitable for arbitrary maneuver trajectories,as well as the dynamic object detection and parameter estimation problems based on this model,and the main contents and innovation points are as follows.1.A multi-channel maneuver trajectory SAR dynamic target complex image domain impulse response signal model is proposed,which establishes the quantitative constraint relationship between the impulse response of the moving target under any maneuver trajectory and the radar system parameters,scene configuration and dynamic target parameters,which lays an important theoretical foundation for the maneuver trajectory SAR-GMTI technology.Firstly,a single/multi-channel SAR dynamic/static target unified signal model with three-dimensional acceleration components is established,and then the analytical expression of the impulse response of the dynamic target complex image domain is derived based on the time domain BP algorithm.Through theoretical analysis,it is found that according to the different 2-dimensional velocity components of the moving target,it will present four different impulse response types in the complex image domain,namely point shape,straight line shape,surface shape,and slash shape,and different response types have different orientation,distance direction envelope and phase modulation,which are collectively referred to as MTIR signal models in this thesis.The influence of the three-dimensional acceleration component of the platform on the MTIR model is different: the influence of azimuth acceleration is much greater than the effect of the distance acceleration,the impact of the height direction acceleration is the smallest,and when the platform only exists in the height direction acceleration(such as pulling up,dive maneuver trajectory),it can be equivalent to the uniform linear motion.The thesis simulates the impulse response results of the dynamic target complex image domain under four typical trajectories,which are consistent with the theoretical analysis and verify the correctness of the theoretical derivation.2.A two-channel SAR dynamic target parameter estimation method based on graph aid is proposed.This method first combines the DPCA algorithm with the MTIR model to derive the dual-channel MTIR-DPCA dynamic object detection algorithm model.The distance envelope of the model contains the Doppler fuzzy number information of the moving target,the azimuth envelope contains the azimuth velocity information of the moving target,and the phase contains the fuzzy velocity information of the moving target,which can construct 4 types of dynamic target 2-dimensional parameter estimation formulas,so that the dynamic target detection result graph and estimation can be used to achieve rapid estimation of the 2-dimensional speed parameter of the moving target without blind search.Through theoretical model analysis,it can be found that the model does not need azimuth time-shift registration and does not need to meet the DPCA conditions,thus breaking the limitation that the traditional DPCA method cannot be used for maneuvering trajectories.Due to the presence of strong clutter and channel errors,there is a certain error in the estimation results.In order to further improve the estimation accuracy of 2D speed,a refocus model based on MTIR model is constructed by using image entropy minimization criterion for small-scale iterative search.This model can more accurately estimate the 2-dimensional velocity component of the moving target,obtain an estimate of the true position of the moving target and achieve the refocus of the moving target.The simulation results of typical maneuvering trajectories demonstrate the effectiveness of the proposed method.3.A blind speed estimation method for dynamic targets of curved trajectory twochannel SAR is proposed.Aiming at the blind speed problem of the traditional DPCA method,this thesis proposes a blind speed estimation method for the blind speed estimation method of the two-channel SAR moving target by using the constraint relationship between the dynamic target complex image domain impulse response and the radar system parameters,the scene configuration and the dynamic target parameters of the radar system when the platform turns,so that the moving target at the blind speed point can be detected.This method divides the aperture of the curve trajectory platform into two azimuths,and combines the MTIR-DPCA algorithm model to perform a subvisual joint estimation of the 2-dimensional velocity parameters of the moving target to solve the blind speed problem.Simulation experiments show that the proposed method can effectively detect the moving target at the blind speed point,and estimate its 2-dimensional velocity and real position parameters. |
PDF Full Text Request