Study On High Resolution ISAR Imaging And Fine Motion Compensation Techniques

Inverse synthetic aperture radar(ISAR)imaging technique plays an important role in military and civil fields with its unique advantages of all day,all weather and high resolution.ISAR provides a powerful technical support for classification and recognition of non-cooperative targets through high-resolution imaging of space,aerial and marine targets.In order to meet the more and more application requirements,ISAR is developing in the direction of multifunction,multidimension and refinement.The diversity of working modes and system structures,as well as the complexity of targets' motion,greatly challenge the current ISAR imaging algorithms in high-resolution imaging and fine motion compensation.With the support of the 13 th five year plan equipment pre research key project and other projects,this paper mainly focuses on the fine motion compensation,optimal imaging time interval selection,sparse signal processing and three-dimensional(3-D)imaging in ISAR imaging,aiming to enhance the resolution of imaging results and improve the accuracy and robustness of motion compensation,so as to lay a good foundation for subsequent target recognition.The main contents of this paper are as follows:(1)Joint azimuth scaling and range spatial-variant autofocusThe ISAR images obtained by the classical Range-Doppler(RD)algorithm only reflects the Doppler information of the target in the azimuth dimension,rather than its real size information.In order to facilitate the subsequent feature extraction and target recognition,it is necessary to realize the azimuth scaling of the RD images.Hence,a joint azimuth scaling and range spatial-variant autofocus algorithm is proposed in Chapter 2.By solving the minimum entropy optimization problem,the optimal estimation of the effective rotational velocity(ERV)can be obtained.On this basis,the range spatial-variant autofocus is completed while the azimuth scaling achieved,so the focusing performance of ISAR images is further improved.(2)Joint translational motion compensation and fine azimuth scalingBy analyzing the homology of range shift and initial phase error,a joint translational motion compensation algorithm is proposed in Chapter 3.Moreover,aiming at the distortion problem of azimuth scaling caused by the inconsistency between the geometry center and the equivalent rotational center of the target,the joint equivalent rotational center shift value and ERV signal model is established.By solving the maximum contrast optimization problem,the optimal estimations of translational motion parameters,equivalent rotational center shift value and ERV can be achieved jointly,so that the joint translational motion compensation and fine azimuth scaling can be realized robustly and accurately even at a low SNR.It should be pointed out that the “joint” here has two meanings: one is the joint correction of range shift and initial phase error caused by translational motion;the other is the joint realization of translational motion compensation and azimuth scaling.(3)Optimal imaging time interval selection for marine targets ISAR imaging The marine target is affected by wave disturbance so has strong maneuverability.Therefore,it is necessary to select the optimal imaging time interval.In order to solve this problem,we propose an optimal imaging time interval selection technique for marine targets ISAR imaging based on sea dynamic prior information in Chapter 4.Instead of the data-driven processing,the proposed algorithm uses the sea dynamic prior information(such as sea state,wave direction angle and navigation speed)to directly derive the analytical expressions of 3-D swaying motions of marine targets according to the principle of hydrodynamics.Then,the ERV vector of the target is derived based on the geometry relationship between the translational and rotational motion of the target.Based on the ERV vector and maximum contrast criterion,the optimal imaging time interval is determined accurately.The proposed algorithm avoids the interference of noise and clutter in the echo signals,and does not need the imaging processing by sliding window,so it has the advantages of strong robustness and small calculation amount.(4)Fine phase autofocus for maneuvering targetsFor maneuvering targets,not only the translational motion will produce phase errors,but also the rotational motion will cause the images blurring.Therefore,a fine phase autofocus algorithm is proposed in Chapter 5.A two-dimensional(2-D)spatial-variant phase error model is established.The model not only considers the time-varying characteristics of targets' translational velocity and ERV,but also takes the rotational property of imaging projection plane(IPP)into account.A two-step estimation method including coarse estimation and fine estimation is used to obtain the optimal estimations of the model parameters,so as to realize the fine phase autofocus of maneuvering targets.The effectiveness and necessity of the proposed algorithm are verified by simulated and real data experiments.(5)High-resolution ISAR imaging and motion compensation of 2-D sparse data Aiming at the sparse stepped frequency modulation and sparse aperture waveform(SSFM-SAW)echo signals received by the multi-function radar system,Chapter 6 of this paper proposes a high-resolution ISAR imaging and motion compensation algorithm with2-D joint sparse reconstruction(2D-JSR).The proposed algorithm constructs a 2D-JSR dictionary,and describes the translational and rotational motion errors as model errors,thereby establishing a SSFM-SAW signal model.Based on the Bayesian compressive sensing(BCS)theory,the sparsity-driven optimization problem is derived according to the SSFM-SAW signal model,and the improved quasi-Newton method is used to efficiently solve it,thereby jointly realizing 2-D joint sparse reconstruction and motion compensation.The proposed algorithm makes full use of the 2-D coupling information of the echo signals,avoids the error transmission between 2-D cascaded sparse reconstruction,and realizes the joint translational motion compensation and range spatial-variant autofocus,so it has the edges of high accuracy and strong robustness.(6)Interferometric ISAR(In ISAR)imaging of maneuvering targets with 2D-JSR Aiming at the maneuvering targets with 3-D rotational motion under the sparse frequency bandwidth and sparse aperture(SFB-SA)signals,Chapter 7 presents an In ISAR imaging algorithm of 3-D maneuvering motion targets based on 2D-JSR.The proposed algorithm derives the analytical expressions of non-spatial-variant wave path difference(NSVWPD)and spatial-variant wave path difference(SVWPD)by analyzing the quantitative relationship between the target's motion and spatial position.Among them,the SVWPD caused by the 3-D rotational motion is a new concept we put forward in this article.It is not only a function of slow time,but also has spatial-variant property.The experimental results show that for the maneuvering targets with 3-D rotational motion,the compensation of SVWPD is indispensable,which directly affects the accuracy of In ISAR imaging.Based on this,we propose a joint wave path difference compensation algorithm(JWPDC),which not only realizes the image registration by combining multiple channels,but also jointly compensates NSVWPD and SVWPD,so the fine image registration can be realized.For SFB-SA signals,in order to retain the high correlation between interferometric phase information of different channels,we develop a joint multi-channel 2D-JSR(JMC-2D-JSR)algorithm.In addition,for the motion errors caused by the translational motion and 3-D rotational motion,we use the algorithms proposed in Chapter 3 and Chapter 5 to achieve precise motion compensation.Through the iterative processing of JMC-2D-JSR and JWPDC as well as the motion compensation algorithms,the high-accuracy 3-D geometry reconstruction results of maneuvering targets with 3-D rotational motion can be obtained.