Study On Imaging Algorithm And Motion Compensation For High-speed Manoeuvring-platform SAR

Synthetic Aperture Radar (SAR) is a microwave remote sensing technology withthe capability of working all day and all weather. Due to the special advantages ofhigh-speed maneuvering-platform SAR, it has got growing interests in recent years. Onone hand, the detective ability of all day and all weather for the maneuvering-platformcan be improved by SAR system; on the other hand, the maneuvering-platform hasmore freedom with SAR system equipped in complex environments. The highresolution images with physiognomy embedded in are obtained by SAR, and they arethen matched with the reference images stored in the database, the real time position ofitself is subsquently calculated by the geometry relationship to correct the INScurriculum errors in order to improve the precision of orientation. Besides, thetwo-dimensional high-resolution image and the scattering information of importantregions can be directly obtained by SAR equipped on the platform, based on which thefollowing feature extraction, classification and recognition process procedures can becarried on.However, unlike traditional airborne SAR or spaceborne SAR, the movingtrajectory of the high-speed maneuvering platform is more complicated, which isleading to difficulties in directly applying available imaging algorithms for traditionalSAR into it. Thus, this paper mainly focuses on the problems of restriction of azimuthfocusing depth, high-resolution imaging algorithms and also the imaging algorithm forFMCW. Moreover, to further improve ability of concealment, counter-concealment andscattering information capture, accommodating the bistatic configuration into thehigh-speed maneuvering-platform SAR has been a new trend. Hence, motioncompensation for bistatic SAR has also been studied. The main research achievementsare given as follows.1. The issue of restriction of azimuth focusing depth in high-speedmaneuvering-platform SAR is studied. Owing to the downward movement ofmaneuvering platform with high vertical velocity and high acceleration, the assumptionof translational invariance in azimuth is no longer valid, which makes it difficult forSAR imaging processing. To address this problem, an imaging algorithm based onazimuth nonLinear chirp scaling (NLCS) for high-speed maneuvering-platform SAR isproposed. After range cell migration correction and range compression, the Dopplerrates of echo signal are equalized via the operation of azimuth NLCS, and the focusing depth and the focusing quality are effectively improved.2. A full-aperture based imaging algorithm for high-speed maneuvering-platformSAR is studied. Owing to the large diving velocity and acceleration in the downwardmovement of maneuvering platform, the accurate expression of2-dimensional (2-D)spectrum of SAR echoes is difficult to be derived, which makes it difficult for SARimaging processing. To solve this problem, an imaging algorithm for high-speedmaneuvering-platform SAR based on the model of hyperbolic range equation withlinear modifying is proposed. By simplifying the range history with a hyperbolic rangeexpression and incorporating a linear modifying component, the2-D spectrum for SARechoes can be readily derived via the principle of stationary phase (POSP), based onwhich the high-efficient frequency-domain SAR algorithm can be well designed. In theproposed algorithm, the derivation of2-D spectrum can be simplified and theexpression of the2-D spectrum is brief and concise with clearer meanings, which isconvenient for signal analyses and imaging process procedures. Results from pointtargets simulation are provided to validate the effectiveness of the proposal.3. The imaging process procedures for high-speed maneuvering-platformFMCW-SAR are studied. According to the downward movement of high-speed andmaneuvering platform, the signal model of FMCW-SAR is established. Based on thesignal model analyses, it can be found that the effects of quadratic coupling phase arepeculiarly introduced by the continuous motion under the condition of high-speed andmaneuvering platform, which is distinguished from traditional airborne FMCW-SARand adversely affects the focusing in range direction. To address this problem, anFMCW-SAR imaging approach is proposed, in which the quadratic coupling phase termare appropriately compensated in2-D frequency domain and the focusing quality issubsequently improved. Simulation results of point targets are provided to validate theeffectiveness of the proposal.4. Motion compensation for bistatic SAR (BiSAR) with high-speed maneuveringplatform is studied. In BiSAR imaging processing, not only motion error of movingplatform, but also inaccuracy of data acquisition geometry will introduce additionalphase error that seriously degrades focusing quality of the final image. Based on thesignal model established according to the BiSAR configuration, the influence of theinaccurate geometry information is analyzed in detail and an applicable motion compensation approach is proposed. By using Doppler rates estimation, the phase errorsfrom both motion error and inaccurate geometry are appropriately compensated, and thefocusing quality of the final image is consequently improved. Analyses of acquired rawdata are presented to verify the effectiveness of the proposal.5. In the configuration of large bistatic angle and the condition of signal propertieswith intense variance in azimuth, motion compensation for bistatic high-speedmaneuvering-platform SAR is studied. In bistatic SAR with high-speed maneuveringplatform, the assumption of translational invariance is no longer valid. After the rangecell migration correction (RCMC), range compressed signal in the same range gateexhibits azimuth-variant Doppler rates that make it difficult for the phase error to beestimated from the echoes by using the available phase gradient autofocus (PGA)algorithms. To deal with this problem, a modified PGA algorithm is proposed.Comparing with the traditional PGA, the residual quadratic phase (RQP) caused by theazimuth-variant Doppler rates is additionally estimated and compensated. As theinfluence of the azimuth-variant Doppler rates is greatly reduced, a phase gradientestimator is subsequently applied for accurate phase error retrieval. Analyses ofacquired raw data are presented to verify the advantages of the proposal.