Research On Real-time Processing Techniques For FMCW SAR

Benefiting from the features of small size,light weight and low power dissipation,frequency modulated continuous wave synthetic aperture radar(FMCW SAR)has been a research hotspot in the last decade.As we all know,the real-time processing technique is absolutely necessary for SAR system,which could immensely improve the value of SAR application.However,the SAR real-time processing has the characteristics of huge data volume and heavy computational load.Moreover,the limited size,weight and power dissipation of FMCW SAR will severely challenge the realization of high resolution real-time mini-SAR imaging.This dissertation focuses on the real-time processing techniques for FMCW SAR,which contains real-time imaging algorithms,motion compensation,real-time parallel processing technique,real-time SAR Video forming technique and the feasibility of real-time processing technique.Combining the theoretical investigation with engineering practice,beneficial results have been achieved.The main works and achievements of this dissertation are summarized as follow:1.The real-time imaging algorithm and motion compensation of FMCW SAR are studied.The main work includes:(1)The FMCW SAR imaging geometric under different conditions are studied,and the influence of intra-pulse movement is analyzed in detail.(2)The imaging algorithms for FMCW SAR are contrastive analyzed,and the motion compensation method for real-time imaging is presented.(3)Considering the feasibility,calculation and processing accuracy of different imaging algorithms,the FMCW SAR real-time processing flow is presented.2.The parallel processing technique for mini-SAR real-time imaging is studied.The main work includes:(1)Based on the miniature hardware platform for real-time processing,the size of the real-time data frame is analized and the the computation burden of the real-time imaging is evaluated in detail.(2)A fast corner turning method is proposed and the time-consuming corner turning operation could be realized effectively.(3)An efficient parallel processing architecture based on multi-core digital signal processor(DSP)is proposed,which could efficiently realize the algorithm mapping in parallel.(4)A data playback platform is established to increase the efficiency of the algrothm mapping,and the real-time ability is tested and verified by the playback system.3.The real-time SAR Video forming technique is studied.The main work includes:(1)The distortion between SAR real-time images is analyzed in detail.(2)A new distortion correction approach for FMCW SAR real-time imaging is proposed,which could effectively reduce the time consumption from 0.3 s to 0.717 ms.(3)The residual offset error between the rotated sub-images is derived and compensated with the first-step motion compensation(MOCO).(3)To achieve consecutive SAR Video,an effective approach is proposed to eliminate the dislocation and defocusing in the distortion corrected real-time images.(4)The simulation and airborne experiments have been performed to verify the feasibility and efficiency of the proposed methods.4.The feasibility of real-time processing technique on different platforms is studied.The main work includes:(1)The motion characteristics of light airplane,sports utility vehicle(SUV)and multi-rotor unmanned aerial vehicle(UAV)platform are comparatively analyzed.(2)The real-time motion compensation under different platforms are studied analyzed respectively.(3)A fast method based on the decimation in time is proposed,which is conducive to realize the mini-SAR real-time imaging under multi-rotor UAV platform.(4)The processing of the measured data has verified the validity of the proposed approaches.This dissertation has made contributions both in theoretical investigation and engineering practice.The achieved research results have been applied to the mini-SAR real-time processing system,and the achievements have important implications for FMCW SAR development.