Research On Deceptive Jamming And Evaluation Methods Against Synthetic Aperture Radar

Synthetic aperture radar(SAR)is an effective means of earth observation,since it can provide two-dimensional(2-D)high-resolution images under any meteorological and illuminative conditions.These properties result in SAR's wide use in civilian applications,such as terrain mapping,oceanographic observation,disaster prediction,and crop assessment.In military applications,SAR is a vital tool for regional reconnaissance,precision guidance and situational awareness.Because of their broad application,SAR jamming technology has gained more and more attention in the field of electronic counter measure(ECM).Due to its low jamming power demand,strong concealment and high fidelity of jamming image,deceptive jamming has become the focus of current ECM research.At the same time,SAR deceptive jamming also faces some technical bottlenecks and challenges.First of all,with the emergence of large scene that are controlled by new SAR working mode with flexibly beam scanning and high-resolution imaging capbaility,SAR deceptive jamming faces the challenge of wide-area complex Doppler modulation.Secondly,the accuracy of SAR jamming depends heavily on the accuracy of reconnaissance parameters.It is necessary to study high-fidelity deceptive jamming methods with limited reconnaissance accuracy.Finally,the performance evaluation of SAR deceptive jamming is an important part of SAR jamming,and there is no effective system evaluation method at present.Under the support of the projects of the 13 th five year plan “Countermeasure Technology of XX synthetic aperture radar Image Processing”,“XX inversion technology of spaceborne synthetic aperture radar”,GF innovation special zone project “XXX complex agile electromagnetic signal recognition technology” and the crosswise project “Research of XXX radar jamming evaluation”,this dissertation focuses on the key problems of SAR deceptive jamming and performance evaluation,and puts forward the corresponding solutions.The main work and innovations of this dissertation are carried out from the following aspects: 1.Firstly,the imaging geometric model skethes of single station SAR with different working modes are established,and the basic processing flow of SAR imaging formation is discussed.Then,the mechanism of traditional deceptive jamming in stripmap SAR is analyzed in detail,and the generation model of deceptive jamming signal is established.Meanwhile,the power requirements of deceptive jamming and barrage jamming are compared,which lays a theoretical foundation for the optimization design of subsequent deceptive jamming algorithm and the construction and implementation of evaluation model.2.To deal with the challenge of Doppler complex modulation of wide-scan large area deceptive jamming caused by2-D flexible beam scanning of SAR working in a new mode: terrain observation by progressive scans(TOPS),a fast deceptive jamming method against TOPSAR based on deceptive jamming template partioning is proposed.Firstly,the characteristics of TOPSAR are analyzed,thereby generating flexible false targets with varying Doppler centroids.This method can solve the problem of Doppler phase history variation in range effectively.Furthermore,the jamming problem is decomposed into simpler sub-problems of the same type,according to the relative position between the jammer and the TOPSAR satellite.The partial deceptive jamming modulation function can be pre-generated before the radar parameter reconnaissance.The partitioned scheme produces multiple deceptive jamming batches in parallel on the TOPSAR image to protect large regions of interest(ROIs).Therefore,the real time calculation load can be reduced.Finally,the effectiveness and flexibility of the proposed scheme is provied by both simulated scatterer data and measured scene data experiments,where the details are presented by point scatterers jamming,and the overall effect is presented by scene jamming.3.In order to solve the problem of defocusing of false targets in SAR image due to the insufficient precision of the detection parameters in the existing SAR deceptive jamming methods,we propose a new deceptive jamming method using networked receivers to perform deceptive jamming against multi-mode near-field SAR.Use multiple receivers to measure the time difference of beam arrival(TDOA),we solve the instantaneous modulation parameters precisely via spherical intersection(SX)method.We then solve the time delays and frequency offsets of the deceptive signal without approximation to avoid accumulating electronic reconnaissance errors with traditional deceptive methods.The accuracy and stability of the proposed deceptive jamming method is relatively high,which can be extended to the jamming of multi-mode SAR.Additionally,we design the distribution of receivers to minimize the condition number,reducing our method's sensitivity to measurement error.Finally,the simulation results prove that the method is robust to deceptive jamming.Moreover,the accuracy and stability of the proposed deceptive jamming method are relatively high,which can be extended to the jamming of multi-mode SAR.4.In view of the single evaluation index and poor universality of the existing SAR active deceptive jamming evaluation methods,we propose a new method of performance evaluation of deceptive jamming based on the combination of typical explicit deceptive jamming evaluation indicators and recessive indicators extracted by convolutional neural networks(CNNs).This proposed method first analyzes the applicability of the existing explicit evaluation indicators with clear physical meaning,and then uses CNNs to extract the recessive indicators that can reflect the in-depth feature of the deceptive jamming samples.Then,the jamming evaluation indicators obtained by the two parties are obtained.We obtain the level of deceptive jamming through the use of the Softmax activation function in a fully connected network.Moreover,we analyze the influence of SAR platform track measurement errors on deceptive jamming,and introduce different measurement errors to generate different levels of deceptive images,thereby solving the problem of no enough deceptive samples.Finally,the moving and stationary target acquisition and recognition(MSTAR)database is used as the deceptive jamming template to verify the effectiveness of the proposed method.