Study On ISAR Ultrahigh-Resolution Imaging Techniques

Radar imaging technology has the battlefield advantages of ultra-long-distance detection and working around the clock.With the improvement of national economic strength and the trend of future battlefield informatization,radar imaging technology plays a vital role in mapping for civil use and battlefield reconnaissance.Among them,inverse synthetic aperture radar(ISAR)is the main means of imaging a single non-cooperative target,and synthetic aperture radar(SAR)is mainly used for ground mapping of large scenes.In order to meet the ever-increasing demands for ultrahigh-resolution(UHR)imaging and tiny component extraction,ISAR hardware devices are developing towards the ability to transmit signals with larger bandwidth and higher carrier frequency.The increase in signal bandwidth and carrier frequency has brought many difficulties to imaging processing,such as insufficient accuracy of traditional imaging models,failure of translational compensation methods,lack of rotation compensation methods,and failure of parameter estimation methods.With the support of the instructor's multiple horizontal and vertical projects,this article focuses on the reserch of high-precision translation compensation method,spatialvariant high-order rotation compensation method,stepped frequency large-bandwidth synthesis method and laser SAR/ISAR fusion imaging.Research on methods and other issues is aimed at improving the imaging accuracy of ultrahigh-resolution ISAR and developing stable parameter estimation algorithms to improve the imaging capabilities of ultrahigh-resolution radars.The research content mainly includes:(1)High-precision translational compensation for ultrahigh-resolution ISARIn recent years,due to the emergence of the ultrahigh-resolution radar system,the working band of the radar has been increasing,and the bandwidth has also increased.In the actual processing of UHR-ISAR imaging,it has been found that the accuracy of traditional translational compensation algorithms is insufficient.The reason is that in the signal processing of large bandwidth,the target rotation movement often produces range cell migration(RCM),which makes the time delay of pulse spatial-variant,and the traditional envelope alignment method that ignores the high-order rotation in the model has a significant decrease in the effect.In addition,due to the increase of the signal carrier frequency,the translational motion information extracted from the envelope cannot meet the wavelengthlevel compensation accuracy,and the unaligned envelope also makes the autofocus method ineffective.The envelope and phase compensation problems make the traditional translation compensation process insufficient in processing large bandwidth and high carrier frequency signals.Therefore,the second chapter of this article first discusses the basic principles of ISAR imaging,and on this basis,further proposes a high-precision translational compensation method based on power gentleness optimization(PGO).This method starts from the idea of inverting the prominent point phase(PPP)for global compensation,so that the translational compensation accuracy reaches the wavelength-level.Based on the separated modeling of the envelope and phase proposed in this paper,the algorithm constructs the optimal criterion of power gentleness and optimizes the solution to ensure the complete extraction of the distinctive point energy.The criterion is composed of normalized amplitude variance and energy constraint parameters,and the compensation vector is obtained by using the order coordinate descent(OCD)method to obtain the compensation vector,and then the complete distinctive point signal is extracted.Then,the phase of the extracted special point signal is extracted by sliding window,and the original data is compensated in the two-dimensional data domain through the special point phase inversion compensation matrix.The results show that the compensation effect of the proposed algorithm is significantly improved compared with the traditional algorithm,and it has higher accuracy in the translational compensation of ultrahigh-resolution ISAR imaging.(2)Ultrahigh-resolution ISAR imaging method combining rotation parameter estimation and high-order motion compensationIn the ultrahigh-resolution non-cooperative target ISAR imaging processing,the highorder RCM and high-order phase errors caused by rotational motion must be accurately compensated.The improvement of the range resolution makes the traditional method that only considers the overall migration caused by translation and the first-order RCM caused by rotation is not accurate enough.The high-order RCM error with unknown parameter is spatial-variant,which also makes the traditional method ineffective.In addition,under the premise of the existence of high-order RCM,the high-order residual phase error composed of the rotation speed ERV and the range center deviation(RCD)is difficult to estimate.Therefore,this paper proposes an ultrahigh-resolution imaging algorithm that combines rotation estimation and high-order motion compensation.On the basis of proposing a parameterized imaging model,this paper establishes the relationship between high-order motion and signal envelope characteristics through display expressions,and further proposes the average range profile sharpness maximization(ARPSM)algorithm to estimate target equivalent rotation speed ERV,and the second-order RCM correction is completed at the same time,so that the two-dimensional signal is decoupled,and the unknown parameters ERV and RCD are also uncoupled.At this time,the rotation estimation can be decomposed into two one-dimensional estimations.This method combines non-uniform fast Fourier transform(NUFFT)to improve the computational efficiency.After completing the ARPSM correction,this paper proposes the entire image sharpness maximization(EISM)algorithm to estimate the remaining unknown parameters RCD and achieve the phase error compensation.Two-step cascade optimization establishes a new ultrahigh-resolution ISAR imaging framework.Both simulation and measured data show the effectiveness of the method proposed in this paper.(3)Ultrahigh-resolution ISAR stepped-frequency synthesis imaging method combined with coherence recoveryAlthough ISAR imaging is developing in the direction of signal processing technology with larger bandwidth and higher frequency bands,the radar system,as the front-end basis,cannot transmit signals with relatively large bandwidth due to the limitations of radar components,which has certain restrictions on the generation of large bandwidth signals.The cost of introducing a new radar system is relatively high,so stepped frequency waveform(SFW)is currently a common method to solve this problem.The prerequisite for obtaining an ideal large-bandwidth signal is to use signal processing algorithms to accurately compensate the errors within and between sub-bands.Regarding the subband error,its main source is the high-speed movement of the target itself and the change of the radar reference signal.For the inter-sub-band error,the possible sources are the phase centers of the two subbands,the position of the signal source,the initial phase difference of the system between the sub-bands,and the target scattering intensity difference.Traditional synthesis methods all need to use overlapping sub-bands to estimate errors,and have low accuracy and poor applicability.In view of the above problems,this paper proposes an ISAR stepped frequency synthesis method combined with coherence recovery.This method is also suitable for Dechirp stepped-frequency signal synthesis with or without overlapping bandwidth.Its synthesis accuracy meets the requirements of imaging,and a complete ultrahigh-resolution ISAR stepped-frequency synthesis imaging process is proposed.For the sub-band error compensation,the method in this paper combines the Dechirp incoherent model and the high-speed target signal model,and fully derives the incoherent expression combining the reference slant distance change and the high-speed motion when Dechirp is received,and the two factors are jointly compensated.Furthermore,this paper proposes a method for estimating the inter-sub-band phase error based on the minimum entropy of high range resolution profile(HRRP).This method uses envelope correlation and side-lobe balance methods to obtain the initial values of linear and constant phase errors,avoids the optimization solution falling into the local optimal value,and improves the calculation efficiency of the algorithm.Furthermore,taking the minimum entropy value of the synthesized signal as the optimization criterion,the DFP quasi-newton method is used to further accurately solve the linear and constant phase errors.Finally,the least square method is used to adjust the amplitude difference between the sub-bands,and then the sub-bands are synthesized to obtain an ideal large-bandwidth signal.Simulation and actual experiments show the effectiveness of the algorithm proposed in this paper.(4)Laser SAR/ISAR fusion full-aperture imaging methodSynthetic aperture ladar(SAL)is a SAR imaging technology that works in the laser band,and its wavelength is at least four orders of magnitude smaller than that of traditional microwave radar.Due to the special working band,SAL radar has the advantages of higher azimuth resolution,near-optical image characteristics,and strong anti-interference ability.However,the small wavelength also causes difficulties for motion error compensation.Since the laser wavelength is generally smaller than the amplitude of the slight vibration of the carrier,the motion error caused by the vibration of the carrier cannot be ignored in SAL processing.The high-order motion error form and small wavelength make the inertial navigation compensation accuracy insufficient.Furthermore,due to the short laser imaging time,short signal wavelength,high-order motion error and other factors,the signal-to-noise ratio is low and there is no distinctive point,the phase gradient autofocus(PGA)method fails in SAL imaging.Therefore,in view of the motion error estimation problem in SAL imaging,combined with the processing ideas of SAR and ISAR,this paper proposes a motion error estimation method based on minimum entropy autofocus(MEA).After the fullaperture data is Deramp processed,the full-aperture data is divided into overlapping subaperture data according to an empirical formula.Then use the MEA algorithm to estimate the sub-aperture motion error.The feature of SAL subaperture imaging scene is small in accordance with the application conditions of ISAR autofocusing technique,and MEA uses image entropy as the criterion for iterative estimation without being restricted by distinctive points.MEA can accurately focus the sub-aperture data,but it will also shift the image while estimating.The MEA algorithm estimated by the ISAR application background has poor phase sequence continuity,which makes the conventional strip SAR stitching method unsuitable.Therefore,in this paper,the overlapping part of the signal is correlated with strong scattering points,and the local linear component obtained by the correlation is used to compensate the image offset between the sub-apertures,and then the linear part of the sub-aperture error is compensated,so that the error is completed.The matched filter imaging is completed after error compensation for the full aperture.The actual measurement experiment of SAL verifies the effectiveness of the proposed algorithm.