Study On Airborne Stripmap SAR Ground Moving Target Imaging And Airborne CSSAR GMTI

As a tool of microwave remote sensing,synthetic aperture radar(SAR)is characterized by all-day,all-weather,and high-resolution,and has been widely used in both military and civilian applications.The SAR ground moving target indication(GMTI)mode plays a very important role in military reconnaissance and traffic monitoring.It is capable of providing high resolution image of the observation scene while detecting and locating ground moving targets.In recent years,with the increasing requirements from both military and civilian applications,SAR-GMTI systems are required to be able to image ground moving targets with high resolution.However,when SAR-GMTI systems implement the imaging processing on the received data of the observation area,the stationary scene will be finely focused,while moving targets may suffer serious defocusing because their phase histories are different from that of the stationary scene.Airborne circular stripmap SAR(CSSAR),which was developed a few years ago,is a new mode of SAR operation.Airborne CSSAR flies along a circular path with constant altitude,and the antenna points outward its circular path.Airborne CSSAR has the advantages of periodic revisiting and large coverage,and thus is an attractive tool for GMTI applications.However,due to the special trajectory of airborne CSSAR,the GMTI methods that are designed for conventional straight-path SAR cannot be directly applied to airborne CSSAR.This dissertation mainly studies the ground moving target imaging method for airborne high-resolution stripmap SAR systems and the GMTI methods for airborne CSSAR.1.The problem of ground moving target imaging with airborne straight-path stripmap SAR is studied and a ground moving target imaging algorithm is proposed for high-resolution SAR systems.The target's range equation is one of the most important parameters in SAR processing,while in most of the existing SAR ground moving target imaging algorithms,a quadratic Taylor approximation is made to it.This approximation would cause serious azimuth sidelobe asymmetry and azimuth defocusing in the case of high-resolution SAR systems.However,if we make a higher order approximation to the range equation or if we directly use the accurate hyperbolic range equation,a high-dimensional parameter search isneeded in the process of imaging,which may result in a large computation load.In this study,we don't make any approximation to the target's range equation.Instead,we develop an equivalent range equation and make a moving target equivalent to a static target.After such equivalence,the commonly used standard imaging algorithms for the static scene can be adapted to focus moving targets.Moreover,for slowly moving targets with unknown motion parameters and position parameters,only a one-dimensional search is required in the process of imaging.Furthermore,since it doesn't make any approximation to the target's range equation,the proposed algorithm is suitable for high-resolution SAR systems.2.The signal model of a ground moving target for airborne single-channel CSSAR is established and the imaging methods for airborne single-channel CSSAR-GMTI systems are studied.The special trajectory of airborne CSSAR fails the imaging methods of conventional straight-path SAR.In this study,the target's range equation and the target's two-dimensional spectrum are first derived.Then,an efficient imaging algorithm is designed for airborne CSSAR-GMTI systems.This algorithm can simultaneously correct the range cell migrations of the stationary scene and multiple slowly moving targets without a prior knowledge of targets' motion and position parameters,and can finely focus the stationary scene and coarsely focus slowly moving targets.Moreover,it only needs four times of one-dimensional Fourier transforms and two phase multiplications,and thus is very efficient.We also propose a ground moving target imaging algorithm for airborne single-channel CSSAR.It implements range cell migration correction in the two-dimensional frequency domain via phase multiplication,and the azimuth compression is implemented in the range Doppler domain.3.The problem of ground moving target detection with airborne CSSAR is studied and an autofocus-based moving target detection algorithm is proposed.In this study,a detailed mathematical description of a moving target's signal model in the SAR image domain is presented,and the defocusing of the moving target is analyzed.The proposed algorithm first divides the image of interest into patches,and then implements autofocus and moving target detection for each patch separately.Different from traditional autofocus-based moving target detection algorithms,the proposed algorithm utilizes the information of the target's Doppler ambiguity in conducting autofocus and thus can detect targets whose Doppler ambiguity numbers are different from that of the static background.Moreover,thepropose algorithm can solve the problem of large range cell migration,which affects most of autofocus algorithms.4.The issue of ground moving target motion parameter estimation and imaging with an airborne dual-channel CSSAR-GMTI system is investigated.In this study,the ground moving target's range equation for airborne dual-channel CSSAR is established and the target's signal model after the DPCA clutter cancellation is derived.The proposed algorithm first estimates and compensates the target's baseband Doppler center,and then a maximum-contrast-based method is used to estimate the target's Doppler chirp rate and Doppler ambiguity number.The moving target imaging is implemented via phase multiplication in the two-dimensional frequency domain,and the target is focused without azimuth dislocation.In estimating the target's motion parameters,besides the Doppler chirp rate and Doppler center frequency,the information of the target's position in the SAR image is also utilized by the proposed algorithm.