Research On Active Repeater Jamming Suppression And Target Detection For Synthetic Aperture Radar

Synthetic Aperture Radar(SAR)active repeater jamming can form deceptive false targets or obscure real targets in SAR images,destroy the imaging features of targets in SAR images,and then affect the whole process of SAR target detection and identification.This thesis focuses on SAR active repeater jamming suppression and target detection under jamming conditions,and proposes a jamming suppression method based on the image domain and a SAR target detection method based on shadow characteristics under jamming conditions.The effectiveness of the proposed method is verified using the selfbuilt SAR active repeater jamming database.The main contents are as follows:Firstly,the SAR active repeater jamming dataset construction method is proposed.Based on the scarcity of measured SAR jamming data,the research on the dataset construction method is carried out.For the jamming signal,the SAR imaging geometry model is constructed,the generation principles of SAR frequency shift jamming,interrupted sampling repeater jamming and micro motion jamming are analyzed,and simulation experiments are conducted to realize the signal domain simulation of the jamming.For the real target echo,the SAR raw echo simulation algorithm is studied.Based on the advantage that the ωK imaging algorithm does not use any approximation and has higher accuracy,the inverse ωK algorithm is used to complete the echo reconstruction of the SAR raw image.The simulated jamming signal is summed with the original SAR echo signal obtained by reconstruction and then imaged to obtain the SAR image under jamming.This part takes Mini SAR image data as an example to construct a SAR land scene active repeater jamming dataset and takes Radarsat-2 image data as an example to construct a SAR ocean scene active repeater jamming dataset.The dataset contains 15 kinds of SAR active repeater jamming.The above research work provides abundant data samples for studying the changes in target characteristics in SAR images before and after jamming and the impact of jamming on target detection performance.Secondly,a method of SAR active repeater jamming suppression based on the image domain is proposed.The characteristics of SAR active repeater jamming and real manmade targets on SAR images are analyzed to reveal the differences in distribution,shape,and location,and a SAR active repeater jamming detection method based on the image domain is proposed.By constructing the circle operator and the cross operator to detect the candidate point set of the jamming center,using the DBSCAN algorithm to locate the center of the candidate jamming point set,and using the two-level histogram statistics to determine the center position of the false target of each order of jamming,to realize the detection of jamming.On this basis,the jamming detection results are used to invert the key parameters of jamming,construct an imaging geometry model,simulate the generation of SAR jamming images,and propose an active repeater jamming suppression algorithm for SAR based on image cancellation.This part uses the SAR image data in the Mini SAR active repeater jamming database to carry out experiments on the suppression of interrupted sampling repeater jamming in the range direction,interrupted sampling repeater jamming in the azimuth direction,and micro motion jamming to verify the effectiveness of the jamming suppression algorithm.Thirdly,a target detection algorithm under jamming conditions is proposed.Aiming at jamming affecting SAR image quality and destroying the imaging characteristics of the target,the effect of jamming on the detection performance of CA-CFAR,SO-CFAR and GO-CFAR under different parameters is investigated using the Radarsat-2 active repeater jamming database.For target detection under jamming conditions,a target detection algorithm based on shadow characteristics is proposed.The target shadow is detected by the cascaded OTSU algorithm,and research on target-shadow matching characteristics is carried out using the property that the target in SAR images has a shadow while the false target formed by jamming has no shadow.The number of false alarms is significantly reduced by filtering out the detection results that cannot complete the matching,thus suppressing the jamming.The algorithm has better detection performance under active repeater jamming conditions.