Research On Ground And Sea Targets Processing Technology For Spaceborne SAR

The spaceborne synthetic aperture radar(SAR)works at a high platform height and can achieve all-time and all-weather wide-band imaging of the earth's surface.While the spaceborne SAR observes the stationary scene on the earth's surface for a long time and broadly,it will also observe various moving targets,including ground moving targets and sea ship targets.In order to achieve accurate imaging of the scene and the moving target,it is necessary to carry out the coherent accumulation of the target signal.In this dissertation,the shortcomings of the existing algorithms are analyzed.For high processing accuracy and high processing efficiency,the spaceborne SAR long synthetic aperture time large scene imaging algorithm and spaceborne SAR moving targets accurate imaging and motion parameters estimation algorithms are studied from the perspective of engineering application.The main contents of this dissertation are summarized as follows.In Chapter 2,a wide scene imaging algorithm based on Singular Value Decomposition(SVD)for geosynchronous orbit(GEO)SAR is proposed.For long synthetic aperture time,large scene width,and serious signal space-variance of GEO SAR.This algorithm simplifies the expression form of the complex 2-D spectrum in the method of series inversion(MSR)through the characteristic components after SVD.After simplifying the 2-D frequency expression,the imaging processing flow is naturally simplified.The coupling relationship between the range and the Doppler in the range signal can be obtained through SVD.A Stolt interpolation and a range frequency resampling can directly realize the range space-variance correction of scenes over 100 kilometers.This algorithm can realize ultra-large range block imaging,does not need to subdivide multiple range blocks and has high processing efficiency.For the accurate imaging and motion parameters estimation of maneuvering targets in highresolution spaceborne SAR,a maneuvering target processing algorithm based on adaptive polynomial Fourier transform(APFT)is proposed in Chapter 3.In this algorithm,the moving target signal envelope migration is first corrected by the unified migration correction function constructed by the scene parameters and the range walk correction function constructed by the radial velocity estimated by Hough transform(HT).Then,the difficulty of precise focusing of moving targets lies in the estimation of azimuth Doppler parameters.Estimating only the second-order Doppler parameter does not meet the requirements of focusing accuracy.The algorithms for estimating the second-and third-order Doppler parameters usually have the problems of heavy calculation complexity and estimation error transmission.Therefore,from the perspective of improving the estimation efficiency,a second-and third-order Doppler parameters estimation algorithm based on APFT is proposed.This algorithm can simultaneously estimate the second-and third-order Doppler parameters in a 1-D parameter interval and has high computational efficiency.Furthermore,there is no error transfer problem between the two parameters.Both simulation and real data verify the effectiveness of the proposed algorithm.With the further improvement of spaceborne SAR resolution to decimeter and the further increase of the synthetic aperture time,the computational complexity required for moving target processing under ultra-high-resolution conditions is too large,which affects the timeliness of moving target processing.Chapter 4 proposes an ultra-high-resolution spaceborne SAR maneuvering target accurate imaging and motion parameters estimation algorithm based on subaperture image sequence in response to this problem.In order to significantly improve the processing efficiency of moving targets,this method uses two core ideas: one is to use the moving targets' imaging positions in the subaperture images instead of the traditional phase information;the other is to obtain all moving targets' imaging positions at the same time instead of estimating each moving target one by one.This chapter derives the relationship between the moving target's imaging position in the subaperture image sequence and the motion parameters.According to this relationship,all moving targets' motion parameters can be estimated after extracting all moving targets' imaging position sequence in the subaperture image sequence.Full-aperture ultra-high-resolution fine focusing is achieved by constructing a refocusing function in the 2-D wavenumber domain.In this chapter,the problem of insufficient motion parameter estimation accuracy caused by severe defocusing of some fast targets in the subaperture images is also considered,and an iterative subaperture refocusing method for precise estimation of motion parameters is proposed.The coarse estimated motion parameters are used to perform refocusing on the subaperture image sequence of the fast target,and then the subaperture image sequence after the refocusing is used for the fine estimation,which solves the problem of high-precision estimation of the fast target.Both simulation and real data verify the effectiveness of the proposed algorithm.The single-channel spaceborne SAR has only one channel,so its radial velocity estimation ability is limited.Chapter 5 proposes a single-channel spaceborne SAR ground and sea moving targets positioning and radial velocity estimation method to solve this problem.The traditional single-channel method usually uses the Doppler center information of the signal to measure the moving target's velocity first and then uses the relationship between the radial velocity and the imaging position offset to locate the moving target.Nevertheless,the singlechannel system limits the velocity measurement capability.The method proposed in this chapter takes advantage of the fact that the energy center of the envelope of the moving target signal is still its real azimuth position in the presence of radial velocity.Contrary to the traditional method,estimating the energy center of the moving target signal's envelope,the moving target is located first,and then its radial velocity is estimated.The effectiveness of the proposed algorithm is verified by real-measured spaceborne SAR data.