Research On Key Imaging Techniques Of Synthetic Aperture Radar On Medium/Geosynchronous Earth Orbit

Medium-and geosynchronous-earth-orbit(MEO/GEO)synthetic aperture radar(SAR)is significantly better than current low-orbit SAR in terms of time resolution and coverage.As one of the most important candidates for the next generation of spaceborne SAR missions,it has become a new hot topic in SAR comunities around the world.This paper excavated the characteristics and challenges of MEO/GEO SAR in system designing and signal processing,and focuses on the aspects of imaging parameter design,working mode design,development of large scene imaging algorithm,and analyses of ship imaging characteristics.The innovative points of this dissertation is highlighted as follows:1.A resolution calculation and imaging paramter optimizing method for complex imaging geometries is proposed.Resolution is one of the key indicators of a SAR system,which is usually determined by the bandwidth,synthetic aperture angle and imaging geometry.The imaging geometry of the MEO/GEO SAR is complicated due to the time-varying satellite velocity,orientation of the beam pointing,and the earth surface curvature,which leads to the failure of the traditional methods of resolution evaluation and imaging parameter design.This dissertation starts with the range and azimuth resolutions on the slant-range plane,and by deriving the geometric projecting relationship between the resolutions on the slant-range plane and the ground plane,the analytical representation of the ground resolution ellipse is obtained,and the optimizing method of the signal bandwidth and synthetic aperture time is proposed according to the constraint of the long axis and the area maximization criterion.This method has also been extended to the bistatic SAR system.This work offers an efficient analyzing tool to resolution evaluation for complex MEO/GEO SAR imaging geometry,and provide a solution to the complex parameter design.2.A 2-dimensional beam steering(2DBS)method is proposed.Orbit curvature of highaltitude MEO/GEO causes obvious time-varying characteristics in the beam footprint speed,and it is difficult to keep the beam dwell time and azimuth resolution constant.In addition,severe range migration in the large squint mode will bring a great burden to data storage and limit the imaging width.This dissertation proposes a two-dimensional beam scanning method suitable for MEO/GEO SAR to attack these two problems.Through the design of azimuth beam scanning rate,the beam dwell time can be controlled to achieve the required resolution.Combining with the elevation beam steering with variable-pulse-interval technique can effectively reduce the amount of data or expand the imaging width in squint mode.This method provides the theoritical basis for the design of a universal MEO/GEO working mode.3.A joint time-doppler resampling(JTDR)method is proposed.To address the serious azimuth spatial variation of the MEO/GEO SAR signal caused by the time-varying satellite speed and curved trajectory,an azimuth time-domain resampling method is proposed to remove the azimuth spatial variation of the Doppler rate at the reference range;The improved range chirp scaling(CS)algorithm is developed to realize the uniform range migration correction.For the residual spatial variation of the Doppler parameters that inflence the azimuth focusing depth,a second-step Doppler resampling is proposed,with which the residual Doppler spatial variation modulation is uniformly corrected.Then,the signals across the large scene can be well focused.This work provide an concise and accurate imaging algorithm for MEO/GEO SAR imaging in broadside mode.4.A variable-interpulse-time technique(VIPT)and a modified JTDR are proposed for highly-squinted-mode imaging.This dissertation first analyzes the main problems facing MEO/GEO high-squint imaging including ultra-long synthetic aperture time,severe range migration,and unbalanced resolution in different directions;then,the variable-inter-pulsetime technique is proposed to for data transceiving to accomadate the serious range walking at large squint;in addition,the joint time-frequency resampling algorithm is extended to high-squint mode imaging.First,the azimuth time-domain resampling is used to process the azimuth spatial variance of the signal envelope,and then an extended range migration algorithm(RMA)is proposed to deal with the range-variant range migration,and finally the Doppler resampling method is adopted to process the residual high-order Doppler parameters.Since the azimuth-time resampling can directly process the non-uniformly collected echo data,the algorithm has a significant advantage in efficiency over traditional algorithms when processing high-squint non-uniformly sampled data.5.An efficient imaging algorithm is proposed based on optimal coordinate system transformation(OCST).Conventional MEO/GEO SAR imaging algorithms usually deal with the range and azimuth variation of signals separately,which leads to a rather complex processing procedure,this dissertation,instead,proposes an efficient solution to the processing of the two-dimensional spatial variations.Starting from the Doppler rate distribution of the scatterers in a large scene,we propose to transform the signal into the optimal coordinate system,then,the two-dimensional spatial variations are concentrated in one dimension,and the new azimuth dimension corresponds to the Doppler rate contour.In this way,the data satisfies the azimuth-translation-invariance property,and a conventional imaging algorithm that only considers the range spatial variation can be adopted to focus the data.In order to achieve precise focus in the corner area of a large scene,an modified Doppler resampling process is introduced into before the azimuth pulse compression to handle the residual high-order Doppler parameters,and eliminate the azimuth focus position shift caused by OCST operation.This research work provide a new efficient solution to addressing the 2-D spatial variances of the MEO/GEO SAR signals.6.The properties and challenges of long-integration-time ship imaging is analyzed.Widearea ocean target surveillance is one of the most important applications of MEO/GEO SAR,however,due to the slow speed of MEO/GEO SAR satellites,the synthetic aperture time required to obtain azimuth resolution equivalent to the range resolution is greatly increased.When the coherent processing time increases,the complex oscillatory motions of the ship will greatly increase the difficulty of ship focusing.This dissertation first analyzes the timevarying characteristics of the relative motion between the ship and the radar platform;Using the projection slice theorem,the cause of the irregular sidelobes of the point spreading function and the characteristics of the wavenumber domain support are studied;Besides,when the equivalent slant range plane changes,the targets outside the imaging plane cannot be focused and will form complex trajectory on the imaging plane.The cause of the projected trajectory is also discussed.The above analyses provide some theoritical fundamendatals for the developement of long synthetic aperture time ship imaging algorithm.