Accurate correction method for electromagnetic scattering feature flicker of ship image based on MIMO radar

Abstract

The invention relates to a precise correction method for the flicker of electromagnetic scattering characteristics of ship images based on MIMO radar, and relates to the field of correction of flicker of electromagnetic scattering characteristics of ships. The invention aims to solve the problem that the existing imaging results of ship targets have electromagnetic scattering characteristic flicker deviation, which leads to low imaging accuracy of MIMO radar for moving ship targets. The invention comprises: obtaining a time-varying correlation model between the echo of a ship target detected by MIMO radar and the electromagnetic scattering characteristics and positions of a scattering surface element; constructing the time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element and different observation channels and different observation angles of the MIMO radar; constructing a compensation dictionary according to the scattering surface elements visible to the incident wave and the primary reflected wave, filtering the radar echo signal, and then imaging the target using a sparse reconstruction method; and analyzing the micro-Doppler characteristics in the echo at the same time, thereby correcting the RCS offset caused by the movement of the ship in the radar image. The invention is used to eliminate the defocus or distortion of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship.

Description

Technical Field

[0001] The invention relates to the field of ship electromagnetic scattering feature flicker correction, and in particular to a precise correction method for the electromagnetic scattering feature flicker of a ship image based on a MIMO radar. Background Art

[0002] For a long time, the detection and imaging of targets such as ships on the sea surface has been an important research direction in the field of radar applications in ocean detection. The electromagnetic scattering characteristics of ship targets can be intuitively presented in radar images, and the shape, size and other information of the target can be measured from them. However, due to the flickering problem of target electromagnetic scattering characteristics in actual scenes, the radar imaging results of ships will show target distortion, false, and even serious image distortion. In fact, since the target electromagnetic scattering intensity is mainly determined by the radar cross section (RCS) linearly, the primary factor causing the flickering phenomenon of ship electromagnetic scattering characteristics is the change of the ship's RCS in the radar radial direction, which is caused by the continuous change of the incident and reflected electromagnetic vectors of the scattering surface element. In this regard, related research mainly compensates for the flickering phenomenon of electromagnetic scattering characteristics by modeling the target RCS.

[0003] Regarding the target RCS modeling, the literature (Iervolino P, Guida R, Whittaker PA Model for the Backscattering From a Canonical Ship in SAR Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 9, no. 3, pp. 1163-1175, 2016.) analyzed the static and stable RCS characteristics of the target in the study, and carried out modeling and calculation based on the geometrical optics (GO) method, but did not consider the model errors caused by the radar perspective change, scattering surface element occlusion, target motion and other problems. In this regard, the literature (Zhang Jufeng, Feng Dejun, Wang Xuesong, etc. Research on Dynamic RCS Simulation of Radar Targets. Journal of System Simulation, 2005, 04: 65-68.) obtained the change law of target RCS with various motions through the interpolation method, and added the motion analysis results on the basis of the omnidirectional static RCS model, but the interpolation operation is easy to bring large errors, which reduces the accuracy of the RCS model. In related studies (Tang B, Chen H, Huang L, et al. Dynamic radar cross-section characteristic analysis of wind turbine based on scaled model experimental. Turkish Journal of Electrical Engineering & Computer Sciences, vol. 27, pp. 4689-4701, 2019.) and (Wang C, Wang Y, Li S B. Inverse synthetic aperture radar imaging of ship targets with complex motion based on match fourier transform for cubic chirps model. IET Radar Sonar & Navigation, vol. 7, no. 9, pp. 994-1003, 2013.), the motion response function of the target was constructed, and the correlation model between the incident and reflected electromagnetic vectors and motion was established to describe the time-varying characteristics of the target RCS with motion, thereby improving the accuracy of the dynamic RCS model.On this basis, in order to further improve the applicability of the model in high sea condition scenarios, the equivalent electromagnetic flow method, far- and near-field RCS algorithms, and RCS geometric optics methods have been proposed successively, which can effectively model complex scenarios with different emission spectra, different radar polarization modes, and incident electromagnetic waves from different perspectives. At the same time, in related studies (Shi FY, Li ZQ, Zhang M, et al. Analysis and simulation of the Micro-Doppler signature of a ship with a rotating shipborne radar at different observation angles. IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022) and (Zhao Y, Zhang M, Chen H, et al. Radar Scattering From the Composite Ship-Ocean Scene: Doppler Spectrum Analysis Based on the Motion of Six Degrees of Freedom. IEEE Transactions on Antennas and Propagation, vol. 62, no. 8, pp. 4341-4347, 2014.), RCS error factors such as radar observation angle and ship micro-Doppler spectrum were analyzed and studied, and RCS calculation methods such as geometrical optics and physical optics (GO-PO) were proposed, but the imaging results observed by the real radar system were not combined for corresponding compensation. In addition, in actual application scenarios on the sea surface, the electromagnetic reflection signal of the ship target is synthesized by multiple scattering areas distributed on the ship, which are either strong or weak. The radar system needs to accurately know the precise distribution and real-time changes of each scattering area on the ship, which puts higher requirements on the real-time and accuracy of dynamic RCS modeling. For SAR and ISAR systems, the synthetic aperture time limits the real-time estimation of the ship RCS, and under the synthetic aperture mechanism, this deviation will accumulate during the sampling time, further deteriorating the detection imaging results.

[0004] In contrast, MIMO radar has an efficient signal transceiver mechanism, which can realize single or a small number of snapshot imaging, and can quickly analyze and model the changes in electromagnetic scattering characteristics, avoiding the accumulation of errors with sampling time, which is more suitable for actual sea applications. However, the MIMO radar image still shows the deterioration or distortion of the target scattering characteristics. The reason is that: on the one hand, for the radar system, due to the variability of the radar line of sight and the radial differences of different observation channels caused by the platform movement (Wang W, Hu ZY, Huang P. 3-D MIMO radar imaging of ship target with rotational motions. Radioengineering, vol. 28, no. 4, pp. 776-784, 2019.), the effective RCS projection of the ship scattering surface element in the radar radial direction and each transceiver array element observation channel is constantly changing, resulting in certain modeling errors. On the other hand, the irregular offset of electromagnetic scattering caused by the complex movement of the ship in the scene will also cause RCS flickering, so that there is still a deviation between the constructed model and the actual electromagnetic scattering characteristics of the ship, which ultimately leads to distortion of the radar image. Therefore, the current ship target imaging results still have electromagnetic scattering characteristic flicker deviation, which leads to the problem of low imaging accuracy of MIMO radar for moving ship targets. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the existing ship target imaging results still have electromagnetic scattering feature flicker deviation, thereby resulting in low imaging accuracy of MIMO radar for moving ship targets, and proposes an accurate correction method for the electromagnetic scattering feature flicker of ship images based on MIMO radar.

[0006] The specific process of the precise correction method for the electromagnetic scattering characteristic flicker of ship images based on MIMO radar is as follows:

[0007] Step 1: Obtain the ship target echo detected by the MIMO radar;

[0008] The scattering surface element is obtained by meshing the imaging area of ​​the ship model;

[0009] Step 2: Obtain a time-varying correlation model between the ship target echo detected by the MIMO radar and the electromagnetic scattering characteristics and positions of each scattering surface element;

[0010] Step 3: Using the time-varying correlation model obtained in step 2, the time-varying correlation characteristics of electromagnetic scattering of scattering surface elements and different observation channels and different observation angles of MIMO radar are constructed according to RCS analysis;

[0011] The time-varying correlation feature is obtained by using a virtual channel formed by a transmitting and receiving array element to obtain the effective cross-sectional area S of the scattering surface element under the virtual equivalent array element. mnk , and then use S mnk The time-varying correlation model is obtained in step 2 to obtain the time-varying correlation features;

[0012] Step 4: Perform visibility analysis on the scattering surface elements according to the time-varying correlation characteristics obtained in step 3 to obtain scattering surface elements that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements;

[0013] Step 5: construct a compensation dictionary based on the scattering surface elements obtained in step 4 that are visible to the incident electromagnetic wave and the first reflected wave and are not blocked by other surface elements, and filter the radar echo signal to eliminate the echo signal of the first reflected wave in the radar receiving signal and compensate and correct the echo detection deviation, and then use the sparse imaging algorithm to image the ship target to obtain the radar imaging result;

[0014] Step 6: Based on the radar imaging results obtained in step 5, further eliminate the defocus or distortion problem of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship.

[0015] Furthermore, the step 1 of obtaining the ship target echo detected by the MIMO radar includes the following steps:

[0016] Step 1: construct a ship model and divide the imaging area of ​​the ship model into grids, and then divide the divided grids into multiple scattering surface elements;

[0017] Among them, the imaging area of ​​the ship model is divided into grids according to the radar's own resolution;

[0018] Step 1 and 2: Use the physical optics method to obtain the RCS characteristics of the scattering surface element, and obtain the visibility of each scattering surface element to the incident wave according to the RCS characteristics of the scattering surface element, so as to obtain the electromagnetic scattering intensity σ of each scattering surface element at the stationary time t0 k (σ k ∝S k );

[0019] Among them, S k is the projection of the kth scattering surface element with area A on the radar line of sight;

[0020] Step 13: Using the electromagnetic scattering intensity σ of each scattering surface element obtained in step 12 k Get the ship target echo detected by MIMO radar:

[0021]

[0022] Among them, τ kIt represents the time delay of the signal from the transmitting element to the receiving element through the scattering surface element k, where k is the number of the scattering surface element, K is the total number of scattering surface elements, t is the transmitting time series, and s is the transmitting signal.

[0023] Furthermore, the time-varying correlation model between the ship target echo detected by the MIMO radar and the electromagnetic scattering characteristics and positions of each scattering surface element in step 2 is as follows:

[0024]

[0025]

[0026] in, is the radial electromagnetic scattering coefficient of the scattering surface element k in the observation channel composed of the mth transmitting array element and the nlth receiving array element at time t, φ(t) is the radar observation angle, θ(t) is the ship rotation offset angle, R0 represents the reference distance from the radar to the center of the imaging area of ​​the ship model, (P x , P y , P z ) is the three-dimensional coordinate of the center of the ship model imaging area, j is an imaginary unit, d x is the transmit array element spacing, d y d z are the row and column spacings between the receiving array elements, (x k (t), y k (t), z k (t)) is the coordinate of the kth scattering element at time t, λ is the wavelength of the radar signal, t0 is the stationary moment, n is the number of rows of receiving array elements, l is the number of columns of receiving array elements, and j is the imaginary part of the complex number.

[0027] Furthermore, the step 3 uses the time-varying correlation model obtained in step 2 to construct the time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element and different observation channels and different observation angles of the MIMO radar according to RCS analysis, including the following steps:

[0028] Step 31: Get the virtual channel composed of the transmitting and receiving array elements:

[0029] S mk =S k ·cosβ mk (t) (5)

[0030] S nlk =S k ·cosβ nlk (t) (6)

[0031]

[0032]

[0033] Among them, S k is the scattering cross section of the kth scattering element, S k Perpendicular to the LoS direction, m is the number of the transmitting array element, M is the total number of transmitting array elements, S mk For the transmitting element T m The effective cross section, S nlk For the receiving array element R nl The effective cross section, β mk (t) is the mth transmitting element T m The observation angle to the kth scattering surface element, β nlk (t) is the receiving element R nl The observation angle to the kth scattering surface element, LoS represents the unit observation vector of the radar angle, T m k is the mth transmitting element T m The observation vector to the kth scattering element, R nl k is the receiving array element R nl The observation vector to the kth scattering element, R nl is the receiving array element located at the nth row and the lth column of the receiving array, n=1,2,...,N, l=1,2,...,L, N is the total number of rows of the receiving array, L is the total number of columns of the receiving array;

[0034] Step 32: Obtain the effective cross-sectional area S of the scattering surface element under the virtual equivalent array element for the virtual channel formed by the transceiver array element obtained in step 31 mnk :

[0035]

[0036]

[0037] h mnl k=T m k+R nl k (11)

[0038] Among them, h mnl k is the unit vector from the equivalent array element synthesized by the transmitting array element and the receiving array element to the scattering surface element k, represents the observation angle of the kth scattering surface element of the equivalent virtual channel indexed as m, n, l at time t;

[0039] Step 33: S obtained according to step 32 mnk The time-varying correlation characteristics of electromagnetic scattering of scattering surface elements and different observation channels and different observation angles of MIMO radar are obtained by using formula (4).

[0040] Furthermore, the time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element in step 33 and different observation channels and different observation angles of the MIMO radar are as follows:

[0041]

[0042] Furthermore, in step 4, the visibility analysis of the scattering surface element is performed according to the time-varying correlation characteristics obtained in step 3 to obtain the scattering surface element that is visible to the incident electromagnetic wave and the primary reflected wave and is not blocked by other surface elements, as follows:

[0043] ① Perform direct visibility analysis on each scattering surface element to the incident wave: if the normal vector of the scattering surface element satisfies n·LOS≥0, then the scattering surface element is visible to the incident wave; if the normal vector of the scattering surface element does not satisfy n·LOS≥0, then the scattering surface element is in the non-illuminated area of ​​the electromagnetic incident wave;

[0044] ② Occlusion judgment between two scattering surface elements: If and Then both scattering planes are visible to the incident wave; otherwise, the scattering plane in front blocks the visibility of the scattering plane behind to the incident wave; then b The set of scattering elements max{r k′ LOS,k′=1,2,…,k b} determines the unique visible scattering surface element in the direction of the incident wave;

[0045] ③ Determine the visibility of the first reflection wave of the scattering surface element: When the reflection wave vector of the scattering surface element k1 satisfy When the other scattering elements are Is it visible and will it be blocked by other scattering elements in the middle? If other scattering elements are If the k2 surface element is visible and is not blocked by other scattering surface elements in the middle, then the first reflection wave of the k2 surface element to the k1 surface element is visible. If the condition is not met, then the first reflection wave of the k2 surface element to the k1 surface element is not visible.

[0046] Among them, it is judged whether other scattering surface elements are It can be seen that the ① judgment is adopted to judge whether the middle part will be blocked by other scattering surface elements. The ② judgment is adopted. k , They represent the unit normal vectors of the kth, k1th, and k2th scattering surface elements respectively, and LoS represents the unit observation vector of the radar perspective. is the unit vector of the line connecting the scattering centers of the two scattering planes, i is the unit direction vector of the incident wave, ζ is the preset detection threshold, r k′ is the normal vector of the scattering surface element that satisfies the occlusion condition, k bis the total number of scattering surface elements that meet the occlusion condition.

[0047] Furthermore, the step 5 constructs a compensation dictionary based on the scattering surface elements obtained in step 4 that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements, and performs filtering processing on the radar echo signal, eliminates the echo signal of the primary reflected wave in the radar receiving signal, and compensates and corrects the echo detection deviation caused by the radar viewing angle and channel reciprocity, and then uses the sparse imaging algorithm to image the ship target to obtain the radar imaging result, including the following steps:

[0048] Step 51: Use the scattering surface elements obtained in step 4 to build a compensation dictionary and filter the MIMO radar echo signal to construct a grid vector H;

[0049] The scattering surface element obtained in step 4 is a scattering surface element that is visible to the incident electromagnetic wave and the primary reflected wave and is not blocked by other surface elements;

[0050] First, obtain the MIMO radar echo:

[0051]

[0052]

[0053]

[0054] Among them, q is the path number of the first reflected wave, Q is the total number of paths of the first reflected wave, and A MP represents Q multipath reflections to the scattering surface element, is the path length of the qth primary reflection wave, is the real electromagnetic scattering intensity of K scattering surface elements, represents the RCS gain of the multipath signal on the scattering surface element k, is the potential artifact path, d0 is the real target echo, and d MP Indicates multipath echo, is a matrix of NLM×QK dimensions, represents the array flow matrix of the kth face element in the signal under the qth path, represents the array flow matrix of the multipath signal under the qth path, represents the scattering intensity of the kth bin in the multipath signal under the qth path, a 1 , b 1 、c 1 represents the steering vector of the first face element, a k , b k 、c k is the steering vector of the kth face element, a K , b K、c K is the steering vector of the Kth face element, A is the array flow matrix of the real target signal, and M is the total number of transmitting array elements;

[0055] Then, the scattering surface element obtained in step 4 is used to obtain the potential artifact path in the signal transmission and reception process. Thus construct the grid vector H;

[0056] Step 52: Determine the weight of each element in H:

[0057] Use the following formula to find the viewing direction of each scattering surface element: The corresponding artifact coordinates will be The weight at the corresponding artifact coordinate grid is set to 0, and the weight at other grids is set to 1;

[0058]

[0059] in, yes The corresponding artifact coordinates;

[0060] Step 53: Use H after determining the weight of each element in step 52 and formula (13) to obtain the MIMO radar echo, and then construct the compensation dictionary A Comp ;

[0061] Step 54: Process the echo received by the radar based on the compensation dictionary constructed in step 53, and convert formula (13) into the following formula:

[0062]

[0063] Step 5: Use formula (15) to construct a sparse reconstruction optimization problem and use the sparse imaging algorithm to obtain radar imaging results.

[0064] Furthermore, in step 53, the MIMO radar echo is obtained by using H after determining the weight of each element in step 52 and formula (13), and then the compensation dictionary A is constructed. Comp , as follows:

[0065]

[0066] Among them, U represents the total number of grids divided in the three-dimensional scene, and Φ is the scattering deviation coefficient of each grid point in each observation channel, Represents the observation angle of the equivalent virtual channel with index m,n,l to the u-th grid.

[0067] Furthermore, the step 6, based on the radar imaging result obtained in step 5, further eliminates the defocus or distortion of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship, including the following steps:

[0068] Step 61: Obtain the Doppler frequency shift in the MIMO radar echo signal:

[0069]

[0070] Where V = (v x ,v y ,v z ),ω=(ω x ,ω y ,ω z ) represent the three-dimensional linear velocity and angular velocity of the ship target, Represents the three-dimensional coordinates of the scattering center of the scattering surface element, V d It is the composite speed of the target's motion speed in each dimension;

[0071] Step 62: Obtain the scattering surface element position offset ΔR caused by the ship motion on the radar echo signal phase according to the Doppler frequency shift obtained in step 61:

[0072] f d t=2V d t / λ=2ΔR / λ (18)

[0073] Among them, ΔR represents the three-dimensional coordinate offset of the face element Δx, Δy, Δz;

[0074] Step 63: Get the deflection angle of the ship on the sea surface:

[0075] θ i (t) = A i sin(ω i t+γ i ),i=roll,pitch,yaw (19)

[0076] Among them, roll is the rolling mode, pitch is the pitching mode, yaw is the heading mode, i is the rotation mode, A i , γ i ,ω i denote the amplitude, initial phase and angular velocity of the deflection respectively;

[0077] Step 64: Use the deflection angle of the ship on the sea surface obtained in step 63 to obtain the X, Y, Z three-dimensional coordinate system rotation matrix Ω of the ship's Roll, Pitch, and Yaw motions. r (t),Ω p (t),Ω y (t):

[0078]

[0079] Among them, r is the abbreviation of roll, p is the abbreviation of pitch, and y is the abbreviation of yaw;

[0080] Step 65: Ω obtained according to step 64 r (t),Ω p (t),Ω y (t) Get the coordinate offset of the scattering surface element:

[0081]

[0082] Step 66: Eliminate the electromagnetic scattering characteristic distortion of the scattering surface element according to the coordinate offset obtained in step 65.

[0083] Furthermore, the step 66 of eliminating the electromagnetic scattering characteristic distortion of the scattering surface element according to the coordinate offset obtained in step 65 is specifically as follows:

[0084] First, using (19) to (21), we can obtain the ship's deflection angle at each sampling moment;

[0085] Then, the three-dimensional position deflection vector Δr of the scattering center of the scattering surface element is obtained. k =(Δx k ,Δy k ,Δz k ), and according to Δr k The positive or negative value of LOS determines the deflection direction of each scattering surface element;

[0086] Among them, Δr k LOS is positive, the deflection direction is toward the radar platform, Δr k LOS is negative, and the deflection direction is away from the radar platform;

[0087] Finally, the electromagnetic scattering characteristics of each scattering surface element are compensated according to the deflection angle and deflection direction using formula (6), thereby eliminating the distortion of the electromagnetic scattering characteristics of the scattering surface element.

[0088] The beneficial effects of the present invention are:

[0089] The present invention proposes an effective correction strategy for electromagnetic scattering characteristics of ship targets by analyzing and compensating for the relevant factors that cause the distortion of ship RCS. The present invention analyzes the RCS deviation mechanism caused by factors such as MIMO radar observation channel, LoS angle and ship motion, and proposes an RCS correction scheme based on MIMO radar to correct the scattering flicker problem in the image. The present invention focuses on the visibility change of each facet element to the incident electromagnetic wave and the primary reflected wave, proposes a judgment and correction method, and realizes active compensation in different line of sight or observation channels. The present invention characterizes the correlation characteristics of facet RCS with the ship rotation angle and the scattering center position offset by analyzing the micro-Doppler frequency shift characteristics in the echo, and effectively reduces the electromagnetic scattering characteristic offset of the radar image caused by irregular ship motion. The present invention solves the electromagnetic scattering distortion or distortion problem existing in the detection of ships moving on the sea surface by MIMO radar, eliminates the electromagnetic scattering characteristic flicker deviation in the imaging results of ship targets, and improves the imaging accuracy of MIMO radar for moving ship targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1(a) is a diagram of the ship model used in the experiment;

[0091] FIG1( b ) is a top-down imaging result of the MIMO radar on the target model in a static state;

[0092] Figure 1(c) is the side view imaging result of the MIMO radar on the target model in a static state.

[0093] Figure 2(a) is the RCS projection diagram of the scattering surface element at different viewing angles;

[0094] Figure 2(b) is a graph showing the RCS characteristic changes after the scattering surface element is deflected at different angles;

[0095] Figure 3 It is a schematic diagram of the visibility analysis principle of scattering surface elements;

[0096] Figure 4(a) is a Doppler spectrum characteristic analysis result of a destroyer;

[0097] Figure 4(b) is a Doppler spectrum characteristic analysis result of a certain aircraft carrier;

[0098] Figure 5(a)-Figure 5(b) It is the projection diagram of the imaging result of the ship target on the XY plane and XZ plane in the near-top state when no compensation is performed;

[0099] Figure 5(c)-Figure 5(d) It is the compensation result diagram of Figure 5(a) and Figure (b);

[0100] Figure 6(a)-Figure 6(b) It is the projection diagram of the imaging result of the ship target on the XY plane and XZ plane in the near side view state when no compensation is performed;

[0101] Figure 6(c)-Figure 6(d) Yes Figure 6(a) , 6(b) Compensation result diagram of

[0102] Figure 7(a) is a multi-channel RCS loss diagram under different array element spacing;

[0103] FIG7( b ) is a diagram showing the correction result of the scattering intensity of the scattering surface element;

[0104] Figure 8(a)-Figure 8(b) It is the projection diagram of the ship target image with a primary reflection wave on the XY plane and XZ plane;

[0105] Figure 8(c)-Figure 8(d) It is the RCS offset correction result of the primary reflected wave;

[0106] Figure 9(a)-Figure 9(b) It is the projection diagram of the uncompensated moving ship imaging result on the XY plane and XZ plane at t = 6.3s;

[0107] Figure 9(c)-Figure 9(d) It is the RCS correction result diagram of Figure 9(a) and Figure 9(b);

[0108] Figure 10(a)-Figure 10(b) It is the projection diagram of the uncompensated moving ship imaging result on the XY plane and XZ plane at t = 10.1s;

[0109] Figure 10(c)-Figure 10(d) This is the RCS deviation correction result diagram for Figure 10(a) and Figure 10(b). DETAILED DESCRIPTION

[0110] For electrically large targets such as ships, their electromagnetic scattering characteristics are composed of scattering elements at various locations on the hull. The RCS of the target scattering element is an important physical quantity that characterizes the electromagnetic scattering intensity of the scattering element echo. It represents 4π times the ratio of the echo power scattered by the scattering element to the incident wave power density, that is, the RCS of the scattering element is proportional to its electromagnetic scattering energy coefficient. Therefore, the modeling and calculation of the RCS of a stationary ship have been fully studied in the relevant literature, and a more accurate modeling technique has been proposed using the physical optics method and edge diffraction theory. On the basis of the Stratton-Chu integral, based on the high-frequency approximation, far-field approximation and tangent plane approximation characteristics, the single-station RCS calculation of the target can be simplified as follows:

[0111]

[0112] Where K represents the total number of ship scattering elements, λ is the wavelength of the radar signal, and r k represents the distance between the kth scattering surface element and the radar, S kis the effective RCS of the kth scattering surface element with area A projected on the radar line of sight (LoS) φ direction, j is the imaginary part of the complex number, and a is the edge length of the scattering surface element. Therefore, the present invention is explained in combination with a specific implementation method based on the above-mentioned single-station RCS of the target.

[0113] Specific implementation method 1: The specific process of the accurate correction method of electromagnetic scattering characteristic flicker of ship image based on MIMO radar in this implementation is as follows:

[0114] Step 1: Get the ship target echo detected by MIMO radar:

[0115] Step 1: Build a ship model (such as Figure 1(a)-Figure 1(c) ), and divide the imaging area of ​​the ship model into a grid, and then divide the divided grid into a plurality of scattering surface elements;

[0116] Among them, the imaging area of ​​the ship model is divided into grids according to the radar's own resolution;

[0117] Step 1 and 2: Measure the RCS characteristics of the scattering surface element based on the physical optics method, obtain the visibility of each scattering surface element to the incident wave according to the RCS characteristics of the scattering surface element, and thus obtain the electromagnetic scattering intensity σ of each scattering surface element at the stationary time t0 k (σ k ∝S k ):

[0118] Among them, S k is the projection of the kth scattering surface element with area A on the radar line of sight;

[0119] Step 13: Using the electromagnetic scattering intensity σ of each scattering surface element obtained in step 12 k Get the ship target echo detected by MIMO radar:

[0120]

[0121] Among them, τ k represents the time delay of the signal from the transmitter to the receiving element via the scattering surface element k, k is the number of the scattering surface element, K is the total number of scattering surface element numbers, t is the transmission time series, and s is the transmission signal;

[0122] Step 2: Obtain the time-varying correlation model between the ship target echo detected by the MIMO radar and the electromagnetic scattering characteristics and positions of each scattering surface element:

[0123] Combined with previous research (Wang W, Hu ZY, Huang P. 3-D MIMO radar imaging of ship target with rotational motions. Radioengineering, vol. 28, no. 4, pp. 776-784, 2019.), after carrier removal and matched filtering, the time-varying correlation model between the MIMO radar echo and the electromagnetic scattering characteristics and positions of each scattering surface element in the three-dimensional scene can be obtained:

[0124]

[0125]

[0126] in, is the radial electromagnetic scattering coefficient of the scattering surface element k in the observation channel composed of the mth transmitting array element and the nlth receiving array element at time t, φ(t) is the radar observation angle, θ(t) is the ship rotation offset angle, R0 represents the reference distance from the radar to the center of the imaging area of ​​the ship model, (P x , P y , P z ) is the three-dimensional coordinate of the center of the ship model imaging area, d x is the transmit array element spacing, d y ,d z are the row and column spacings between the receiving array elements, (x k (t), y k (t), z k (t)) is the three-dimensional coordinate of the kth scattering surface element at time t, λ is the wavelength of the radar signal, t0 is the static time, n is the number of rows of receiving array elements, l is the number of columns of receiving array elements, and j is the imaginary part of the complex number.

[0127] Since the overall RCS of the ship is composed of the various scattering elements of the hull, the radar imaging technology is used to process the echo and calculate the electromagnetic scattering characteristics of each scattering element, as shown in Figure 1 (b) and (c), which intuitively presents the detection results of the target, from which the target size, shape and other information can be observed. However, in actual application scenarios, due to the flickering phenomenon of the electromagnetic scattering characteristics of the scattering element, the target in the real radar image is distorted or even some scattering element information is lost, as shown in Figure 2 (a). The main reasons for this phenomenon include: on the one hand, the change of the MIMO radar's own viewing angle and the mutual difference of observations in different channels cannot be ignored. There are differences in the RCS characteristics of the scattering elements obtained in any two observation directions LOS1 and LOS2, resulting in a deviation between the target electromagnetic scattering intensity σ in the radar image and the actual situation. On the other hand, the swaying of the ship itself causes the angle between its scattering surface element and LOS1 to change continuously, causing the radial effective RCS of the scattering surface element to change continuously. After the scattering surface element S moves to S', the spatial angle between its normal vector n and the radar perspective LOS1 changes from β1 to β3, resulting in completely different effective RCS projections S1 and S2 of the scattering surface element in the radial direction of LOS1 before and after the movement, causing the electromagnetic scattering characteristics of the scattering surface element at this location in the image to be distorted. As shown in Figure 2(b), the RCS calculation results after the scattering surface element is deflected at different angles have significant differences under different radar perspectives. In addition, as described in related research (Zhao Y, Zhang M, Chen H, et al. Radar Scattering From the Composite Ship-Ocean Scene: Doppler Spectrum Analysis Based on the Motion of Six Degrees of Freedom. IEEE Transactions on Antennas and Propagation, vol. 62, no. 8, pp. 4341-4347, 2014.), the irregular motion of the ship itself, such as Roll, Pitch, Yaw, etc., not only causes the RCS change of the scattering surface element, but also causes the echo signal phase shift during the continuous sampling process of the MIMO radar, that is, produces micro-Doppler frequency shift, which will also cause the radar imaging results to be distorted.

[0128] Step 3: Using the time-varying correlation model obtained in step 2, the time-varying correlation characteristics of electromagnetic scattering of scattering surface elements and different observation channels and different observation angles of MIMO radar are constructed according to RCS analysis, including the following steps:

[0129] Step 31: Get the virtual channel composed of the transmitting and receiving array elements:

[0130] Firstly, different transmit and receive channels are taken into account in the time-varying correlation model and the corresponding array element steering vectors are constructed, which can be applied to carrier platforms with more transmit and receive channels such as MIMO radar. The RCS radial projection obtained in each virtual channel is actually determined by the observation vector of each channel.

[0131] The virtual channels formed by the transceiver array elements are as follows:

[0132] S mk =S k ·cosβ mk (t)

[0133] S nlk =S k ·cosβ nlk (t)

[0134]

[0135]

[0136] Among them, S k is the scattering cross section of the kth scattering element, S k Perpendicular to the LoS direction, m is the number of the transmitting array element, M is the total number of transmitting array elements, S mk For the transmitting element T m The effective cross section, S nlk For the receiving array element R nl The effective cross section, β mk (t) is the mth transmitting element T m The observation angle to the kth scattering surface element, β nlk (t) is the receiving element R nl The observation angle to the kth scattering surface element, LoS represents the unit observation vector of the radar angle, T m k is the mth transmitting element T m The observation vector to the kth scattering element, R nl k represents the receiving array element R nl The observation vector to the kth scattering element, R nl is the receiving array element located at the nth row and the lth column of the receiving array, n=1,2,...,N, l=1,2,...,L, N is the total number of rows of the receiving array, L is the total number of columns of the receiving array;

[0137] It can be concluded that the effective cross-sectional area corresponding to each array element is related to the observation direction of the array element and the radar line of sight direction.

[0138] Step 32: Obtain the effective cross-sectional area S of the scattering surface element under the virtual equivalent array element for the virtual channel formed by the transceiver array element obtained in step 31 mnlk :

[0139]

[0140]

[0141] h mnl k=T m k+R nl k

[0142] Among them, h mnl k is the unit vector from the equivalent array element synthesized by the transmitting array element and the receiving array element to the scattering surface element k, represents the observation angle of the kth surface element of the equivalent virtual channel indexed as m, n, l at time t;

[0143] Step 33: S obtained according to step 32 mnk The actual value of the electromagnetic scattering coefficient is obtained by using formula (4), that is, the time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element and different observation channels and different observation angles of the MIMO radar:

[0144]

[0145] At the same time, for the RCS detection deviation caused by the change of the radar's own viewing angle φ, the inertial navigation system carried by the platform can often be used to timely obtain the position and attitude changes of the radar platform, and update the radar observation vector LoS in the constructed coordinate system.

[0146] Step 4: Perform visibility analysis on the scattering surface elements according to the time-varying correlation characteristics obtained in step 3 to obtain scattering surface elements that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements:

[0147] As the radar viewing angle changes, for ship targets, it is necessary to determine the visibility change of each scattering surface element to the incident wave at that viewing angle, and then actively compensate for the RCS flickering phenomenon of the scattering surface element in the image according to equations (1) and (2). The visibility (or occlusion) determination condition is:

[0148] ① Perform direct visibility analysis on each scattering surface element to the incident wave: if the normal vector of the scattering surface element satisfies n·LOS≥0, then the scattering surface element is visible to the incident wave. If this condition is not met, it means that the surface element is in the non-illuminated area of ​​the electromagnetic incident wave;

[0149] ② Occlusion judgment between two scattering planes: The scattering centers of each scattering plane are judged two by two, and the vector connected by the scattering centers of two scattering planes satisfies And the normal vectors of both satisfy This means that both scattering surfaces are visible to the incident wave, otherwise the former will block the latter's visibility to the incident wave, and when the blocking condition is met, bThe set of scattering surface elements is composed of max{r k′ LOS,k′=1,2,…,k b} operation determines the unique visible scattering surface element in the direction of the incident wave;

[0150] Among them, r k′ is the normal vector of the scattering surface element that satisfies the occlusion condition, k b is the total number of scattering surface elements that meet the occlusion condition.

[0151] ③ Determine the visibility of the first reflection wave of the scattering surface element: The reflection wave vector of the k1 surface element is Then we need to judge whether other scattering elements are Whether it is visible and whether it will be blocked by other scattering surface elements in the middle, the judgment criteria are the same as ①② above. If the k2 surface is visible and is not blocked by other scattering surface elements, the first reflection wave of the k1 surface element from the k2 surface element is visible, that is, the scattering gain brought by the reflection wave of the k1 surface element needs to be considered in the scattering echo of k2. If the condition is not met, it means that the first reflection wave of the k1 surface element from the k2 surface element is not visible.

[0152] Among them, n k , They represent the unit normal vectors of the kth, k1th, and k2th scattering surface elements respectively, and LoS represents the unit observation vector of the radar perspective. is the unit vector of the line connecting the scattering centers of the two scattering surface elements, i is the unit direction vector of the incident wave, Represents the incident wave vector of the scattering surface element, ζ is the preset detection threshold, which is a small positive number.

[0153] If there is no other scattering surface element blocking the secondary reflection process of the primary reflection wave, there are two situations at this time. One is that the distance between the two scattering surface elements is lower than the radar resolution, and the path of the primary reflection wave is shorter, which will mainly affect the RCS gain of the scattered signal of the scattering surface element; the other is that the distance between the two scattering surface elements is large, and the path of the primary reflection wave is longer, which will affect the RCS gain and delay in the scattered echo, especially Figure 3 As shown, artifacts are generated on the extended path of the radar line of sight. In the present invention, the latter is focused on, that is, considering how to eliminate the artifacts generated when the distance between the two scattering surface elements is large.

[0154] Step 5: construct a compensation dictionary based on the scattering surface elements visible to the incident wave or the first reflected wave obtained in step 4 and filter the radar echo signal, eliminate the echo signal of the first reflected wave in the radar receiving signal, compensate and correct the echo detection deviation caused by the radar viewing angle and channel reciprocity, and use the sparse imaging algorithm to process the ship target to obtain the radar imaging result, including the following steps:

[0155] Step 5.1: Construct a compensation dictionary based on the visibility analysis results obtained in step 4 and filter the MIMO radar echo signal to construct a 0,1 weighted grid vector H; the static ship steady RCS model constructed offline can first restore the electromagnetic scattering characteristics of the scattering surface element that was originally blocked in the image based on the visibility analysis results. Then, it is necessary to consider the elimination of the primary reflection wave artifact. After the carrier removal and matched filtering process, the real MIMO radar echo can be expressed as:

[0156]

[0157]

[0158]

[0159] Among them, q is the path number of the first reflected wave, Q is the total number of paths of the first reflected wave, and A MP represents Q multipath reflections to the scattering surface element, is the path length of the qth primary reflection wave, is the real electromagnetic scattering intensity of K scattering surface elements, It represents the RCS gain of the multipath signal on the scattering surface element k, but this gain is actually unknown. is the potential artifact path, d0 is the real target echo, and d MP Indicates multipath echo, is a matrix of NLM×QK dimensions, represents the array flow matrix of the kth face element in the signal under the qth path, represents the array flow matrix (also called array flow matrix) of the multipath signal under the qth path, represents the scattering intensity of the kth bin in the multipath signal under the qth path, a 1 , b 1 、c 1 represents the steering vector of the first face element, a k , b k 、c k is the steering vector of the kth face element, a K , b K 、c K is the steering vector of the Kth face element, A is the array flow matrix of the real target signal, and M is the total number of transmitting array elements.

[0160] Based on the visibility analysis and judgment of the primary reflected wave in ③, the potential artifact path in the signal transmission and reception process can be obtained. This constructs a 0,1 weighted grid vector H;

[0161] Step 52: Combine (7) to further construct the compensation dictionary A Comp :

[0162]

[0163] Among them, each element of H is based on The weighted result of each grid in the detection area, U represents the total number of grids divided in the three-dimensional scene, and Φ is the scattering deviation coefficient of each grid point in each observation channel, Represents the observation angle of the equivalent virtual channel with index m, n, l to the u-th grid;

[0164] Step 53: Determine the weight of each element in H:

[0165] Combine the radar to find the viewing direction of each scattering surface element The corresponding artifact coordinates, that is, formula (9), and then The weight at the corresponding artifact coordinate grid is set to 0, and the weight at other grids is set to 1;

[0166]

[0167] in, yes The corresponding artifact coordinates;

[0168] Step 54: Based on the compensation dictionary in (8), the echo received by the MIMO radar is processed in advance, so that the model in (7) is converted into:

[0169]

[0170] In this model, the relevant echo signals of the first reflection wave in the original dictionary can be filtered out through compensation processing, and the detection deviation caused by the multi-channel reciprocity of the radar can be corrected.

[0171] Step 55: Use formula (10) to construct a sparse reconstruction optimization problem, and use a sparse imaging algorithm to achieve effective imaging of the ship target to obtain radar imaging results.

[0172] Step 6: Based on the radar imaging result obtained in step 5, further eliminating the defocus or distortion of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship, including the following steps:

[0173] Based on the results obtained from radar imaging, it can be seen from the previous analysis that the linear motion of the ship and the rotations such as Roll, Pitch, and Yaw will cause micro-Doppler frequency shift in the radar echo, causing image distortion and irregular deflection of the surface element, resulting in RCS distortion, that is, In order to compensate for the RCS deviation caused by the movement of the ship, it is necessary to analyze the Doppler frequency shift in the echo and obtain the two elements required for compensation. One is the angle of the facet deflection, and the other is the direction of the facet deflection. The former is mainly caused by the shaking of the ship itself, and the latter determines whether the RCS deviation of the ship facet is positive or negative.

[0174] The micro-Doppler frequency of the ship will cause the true phase of the echo to deviate, causing the radar to offset the target position solution. Consider the radial velocity V of the ship d To synthesize its linear motion and three-dimensional rotation, it is necessary to obtain the Doppler shift;

[0175] Step 61: Obtain the Doppler frequency shift in the MIMO radar echo signal:

[0176]

[0177] Where V = (v x ,v y ,v z ),ω=(ω x ,ω y ,ω z ) represent the three-dimensional linear velocity and angular velocity of the ship target, Represents the three-dimensional coordinates of the scattering center of the scattering surface element, V d It is the composite speed of the target's motion speed in each dimension;

[0178] Step 62: According to the Doppler frequency shift obtained in step 61, the influence of the ship motion on the radar echo signal phase can be obtained as described in formula (12). It can be seen that the micro-Doppler frequency shift in the radar sampling process actually causes the bin position offset ΔR in the echo:

[0179] f d t=2V d t / λ=2ΔR / λ (12)

[0180] The position offset ΔR in the formula can actually be intuitively expressed as the three-dimensional coordinate offset Δx, Δy, Δz of the facet. In other words, as long as the three-dimensional coordinate offset of the facet is obtained, the Doppler frequency shift of the ship motion can be obtained. Then, based on (11) and combined with the existing motion estimation algorithm (for example, Z. Hu, W. Wang and F. Dong. MIMO Radar Accurate Imaging and Motion Estimation for 3-D Maneuvering Ship Target. IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-12, 2021.), the ship's motion speed and angular velocity in each dimension can be solved, and then the actual facet deflection angle can be obtained.

[0181] Step 6.3: The deflection angle of the ship on the sea surface satisfies the quasi-sinusoidal time-varying motion characteristics. For two common large ship types, destroyers and aircraft carriers, the Doppler spectrum of the scattering surface element within the sampling time is as follows: Figure 4(a)-Figure 4(b) As shown, it is proved that the Doppler frequency shift caused by the motion of ship targets in real sea scenes cannot be ignored;

[0182] The deflection angle of the ship on the sea surface is as follows:

[0183] θ i (t) = A i sin(ω i t+γ i ),i=roll,pitch,yaw (13)

[0184] Among them, roll is the rolling mode, pitch is the pitching mode, yaw is the heading mode, i is the rotation mode, A i , γ i ,ω i denote the amplitude, initial phase and angular velocity of the deflection respectively;

[0185] Step 64: By setting the reference coordinate space, the Doppler characteristics of the ship's motion are analyzed based on the echo signals obtained under a small number of snapshots. Combined with the ship target motion speed, angular velocity and other parameters, the three-dimensional coordinate offset of the scattering position can be obtained, and the vector change characteristics of the target surface element center and the radar platform center and array element center can be obtained, which can provide a reference for compensating the dynamic RCS change of the target surface element. In order to clarify the center offset of the scattering surface element, it is necessary to use the X, Y, Z three-dimensional coordinate system rotation matrix of the ship's Roll, Pitch, and Yaw motion, that is, the deflection angle obtained in step 63 is used to obtain the X, Y, and Z three-dimensional coordinate system rotation matrix Ω of the ship's Roll, Pitch, and Yaw motion. r(t),Ω p (t),Ω y (t):

[0186]

[0187] Among them, r is the abbreviation of roll, p is the abbreviation of pitch, and y is the abbreviation of yaw;

[0188] Step 65: Ω obtained according to step 64 r (t),Ω p (t),Ω y (t) Get the coordinate offset of the scattering surface element:

[0189]

[0190] Step 66: Eliminate the electromagnetic scattering characteristic distortion of the scattering surface element according to the coordinate offset obtained in step 65:

[0191] First, combining (13) to (15), we can obtain the ship's deflection angle at each sampling moment;

[0192] Then, the three-dimensional position deflection vector Δr of the scattering center of each scattering surface element is obtained. k =(Δx k ,Δy k ,Δz k ), according to Δr k The positive or negative value of LOS determines the deflection direction of each scattering surface element at this time;

[0193] If Δr k If LOS is positive, the deflection direction is toward the radar platform, and if it is negative, the deflection direction is away from the radar platform;

[0194] Finally, the electromagnetic scattering characteristics of each scattering surface element are compensated according to the deflection angle and deflection direction using formula (6), thereby eliminating the distortion of the electromagnetic scattering characteristics of the scattering surface element.

[0195] Example:

[0196] The effect of the present invention can be verified by the following simulation. In this section, the effectiveness of the proposed compensation scheme will be verified from the aspects of the change of LoS perspective that causes the radar image flicker problem, the difference in radial observation of radar multi-channels, the distortion of the primary reflection wave artifact, and the irregular movement of the ship itself. The execution process is shown in Table 3.

[0197] Table 3

[0198]

[0199]

[0200] First, the parameters of the MIMO radar system used in the experimental part are shown in Table 1. Figures 5(a), (b) and 6(a), (b) describe the actual imaging results of the MIMO radar platform for the ship target from the near top view and near side view. The ship model used in the imaging scene is a certain type of destroyer constructed in the previous article, and the size of the scattering surface element is the same as the size of the divided grid. It can be seen from the figure that compared with the real ship electromagnetic scattering image, after the radar LoS perspective changes, the imaging results of the ship target have more serious electromagnetic scattering characteristic distortion and occlusion problems. The main reason is that the visibility change of the scattering surface element caused by the uncompensated radar perspective and the occlusion problem between the scattering surface elements. In contrast, in Figures 5(c), (d) and 6(c), (d), the correction results of the above problems according to the method of the present invention are presented. Through comparison, it can be concluded that, from the perspective of the radar system, based on the correction strategy of this paper, according to the visibility analysis of each scattering surface element to the incident wave at this viewing angle and the occlusion judgment between the scattering surface elements, active compensation for the RCS flicker deviation of the distorted scattering surface elements in the image can be achieved, which can effectively eliminate the deviation and occlusion in the image, so that the radar imaging results can show more realistic target shape, size and other characteristics.

[0201] Figure 7 describes the RCS loss of the scattering plane element caused by the observation vector difference between multiple channels under different array element spacing. As shown in Figure 7(a), with the increase of array element spacing and the number of array elements, the electromagnetic scattering characteristic loss of the scattering plane element caused by the observation vector difference between virtual channels of different array elements increases. In contrast, Figure 7(b) shows the correction results of this part of the RCS loss according to equations (5) and (6). Considering the influence of noise in the echo, in the radar imaging and signal processing process of this paper, the target detection threshold is set to 0.25 after the echo intensity is normalized. It can be seen from Figure 7(b) that the real multi-channel radial difference will cause the electromagnetic scattering intensity of more scattering plane elements to drop below the noise threshold, which will deteriorate the radar detection imaging of ship targets. According to equations (5) and (6), this part of the loss can be significantly compensated, effectively improving the reliability of radar detection.

[0202] Table 1

[0203]

[0204] Figures 8(a) and (b) show the radar imaging results caused by the secondary scattering process between scattering planes containing the primary reflected wave in the echo. Compared with the standard image in Figure 1, the results obviously show more artifact scattering planes, which is consistent with the analysis results in equation (7). To address this problem, the grid weight matrix is ​​constructed in advance according to (8) to (10) during the radar signal processing process and the corresponding filtering is performed before imaging. The imaging compensation results are shown in Figures 8(c) and (d). Most of the artifact scattering points can be effectively eliminated, which can improve the distortion phenomenon in the radar image.

[0205] Secondly, in order to verify the image correction effect of moving ship targets, Table 2 first summarizes the motion characteristics of a selected destroyer under level 5 sea conditions, and presents the corresponding radar imaging results at two random sampling times of 6.3s and 10.1s, as shown in Figures 9(a), (b) and 10(a), (b). It can be seen from the figure that due to the irregular deflection of the ship target, a more serious flicker phenomenon appears in the radar image, and the scattering characteristics of each scattering surface element have a large deviation, and even some scattering surface element information is lost. In this regard, combined with the known ship motion characteristics, the current radar imaging results are effectively compensated based on (13) and (15), the electromagnetic scattering characteristic deviation of each scattering surface element is obtained, and the RCS characteristics of the obscured scattering surface element are restored in combination with visibility analysis and judgment. Finally, the image flicker correction results at these two moments are presented in Figures 9(c), (d) and 10(c), (d). Obviously, the detection and imaging results of MIMO radar for moving ship targets can be effectively improved.

[0206] Table 2

[0207]

Claims

1. A precise correction method for the flicker of electromagnetic scattering characteristics of ship images based on MIMO radar, characterized in that The specific process of the method is: Step 1: Obtain the ship target echo detected by the MIMO radar; Step 2: Obtain a time-varying correlation model between the ship target echo detected by the MIMO radar and the electromagnetic scattering characteristics and positions of each scattering surface element; Step 3: Using the time-varying correlation model obtained in step 2, the time-varying correlation characteristics of electromagnetic scattering of scattering surface elements and different observation channels and different observation angles of MIMO radar are constructed according to RCS analysis; The time-varying correlation feature is obtained by using a virtual channel formed by a transmitting and receiving array element to obtain the effective cross-sectional area S of the scattering surface element under the virtual equivalent array element. mnk , and then use S mnk The time-varying correlation model is obtained in step 2 to obtain the time-varying correlation features; Step 4: Perform visibility analysis on the scattering surface elements according to the time-varying correlation characteristics obtained in step 3 to obtain scattering surface elements that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements; Step 5: construct a compensation dictionary based on the scattering surface elements obtained in step 4 that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements, and filter the radar echo signal to eliminate the echo signal of the primary reflected wave in the radar receiving signal and compensate and correct the echo detection deviation, and then use the sparse imaging algorithm to image the ship target to obtain the radar imaging result; Step 6: Based on the radar imaging results obtained in step 5, further eliminate the defocus or distortion of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship.

2. The accurate correction method for the flicker of electromagnetic scattering characteristics of ship images based on MIMO radar according to claim 1 is characterized by: The step 1 of obtaining the ship target echo detected by the MIMO radar includes the following steps: Step 1: construct a ship model and divide the imaging area of ​​the ship model into grids, and then divide the divided grids into multiple scattering surface elements; Among them, the imaging area of ​​the ship model is divided into grids according to the radar's own resolution; Step 1 and 2: Use the physical optics method to obtain the RCS characteristics of the scattering surface element, and obtain the visibility of each scattering surface element to the incident wave according to the RCS characteristics of the scattering surface element, so as to obtain the electromagnetic scattering intensity σ of each scattering surface element at the stationary time t0 k (σ k ∝S k ); Among them, S k is the projection of the kth scattering surface element with area A on the radar line of sight; Step 13: Using the electromagnetic scattering intensity σ of each scattering surface element obtained in step 12 k Get the ship target echo detected by MIMO radar: Among them, τ k It represents the time delay of the signal from the transmitting element to the receiving element through the scattering surface element k, where k is the number of the scattering surface element, K is the total number of scattering surface elements, t is the transmitting time series, and s is the transmitting signal.

3. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 2 is characterized by: The time-varying correlation model between the ship target echo detected by the MIMO radar and the electromagnetic scattering characteristics and positions of each scattering surface element in step 2 is as follows: in, is the radial electromagnetic scattering coefficient of the scattering surface element k in the observation channel composed of the mth transmitting array element and the nlth receiving array element at time t, φ(t) is the radar observation angle, θ(t) is the ship rotation offset angle, R0 represents the reference distance from the radar to the center of the imaging area of ​​the ship model, (P x , P y , P z ) is the three-dimensional coordinate of the center of the ship model imaging area, j is an imaginary unit, d x is the transmit array element spacing, d y ,d z are the row and column spacings between the receiving array elements, (x k (t), y k (t), z k (t)) is the coordinate of the kth scattering element at time t, λ is the wavelength of the radar signal, t0 is the stationary moment, n is the number of rows of receiving array elements, l is the number of columns of receiving array elements, and j is the imaginary part of the complex number.

4. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 3 is characterized by: The step 3 uses the time-varying correlation model obtained in step 2 to construct the time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element and different observation channels and different observation angles of the MIMO radar according to RCS analysis, including the following steps: Step 31: Get the virtual channel composed of the transmitting and receiving array elements: S mk =S k ·cosβ mk (t) (5) S nlk =S k ·cosβ nlk (t) (6) Among them, S k is the scattering cross section of the kth scattering element, S k Perpendicular to the LoS direction, m is the number of the transmitting array element, M is the total number of transmitting array elements, S mk For the transmitting element T m The effective cross section, S nlk For the receiving array element R nl The effective cross section, β mk (t) is the mth transmitting element T m The observation angle to the kth scattering surface element, β nlk (t) is the receiving element R nl The observation angle to the kth scattering surface element, LoS represents the unit observation vector of the radar angle, T m k is the mth transmitting element T m The observation vector to the kth scattering element, R nl k is the receiving array element R nl The observation vector to the kth scattering element, R nl is the receiving array element located at the nth row and the lth column of the receiving array, n=1,2,...,N, l=1,2,...,L, N is the total number of rows of the receiving array, L is the total number of columns of the receiving array; Step 32: Obtain the effective cross-sectional area S of the scattering surface element under the virtual equivalent array element for the virtual channel formed by the transceiver array element obtained in step 31 mnk : h mnl k=T m k+R nl k (11) Among them, h mnl k is the unit vector from the equivalent array element synthesized by the transmitting array element and the receiving array element to the scattering surface element k, represents the observation angle of the kth scattering surface element of the equivalent virtual channel indexed as m, n, l at time t; Step 33: S obtained according to step 32 mnk The time-varying correlation characteristics of electromagnetic scattering of scattering surface elements and different observation channels and different observation angles of MIMO radar are obtained by using formula (4).

5. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 4 is characterized by: The time-varying correlation characteristics of the electromagnetic scattering of the scattering surface element in step 33 and different observation channels and different observation angles of the MIMO radar are as follows:

6. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 5 is characterized by: The step 4 performs visibility analysis on the scattering surface element according to the time-varying correlation characteristics obtained in step 3 to obtain the scattering surface element that is visible to the incident electromagnetic wave and the primary reflected wave and is not blocked by other surface elements, as follows: ① Perform direct visibility analysis on each scattering surface element to the incident wave: if the normal vector of the scattering surface element satisfies n·LOS≥0, then the scattering surface element is visible to the incident wave; if the normal vector of the scattering surface element does not satisfy n·LOS≥0, then the scattering surface element is in the non-illuminated area of ​​the electromagnetic incident wave; ② Occlusion judgment between two scattering surface elements: If and Then both scattering planes are visible to the incident wave; otherwise, the scattering plane in front blocks the visibility of the scattering plane behind to the incident wave; then b The set of scattering elements max{r k′ LOS,k′=1,2,…,k b } determines the unique visible scattering surface element in the direction of the incident wave; ③ Determine the visibility of the first reflection wave of the scattering surface element: When the reflection wave vector of the scattering surface element k1 satisfy When the other scattering elements are Is it visible and will it be blocked by other scattering elements in the middle? If other scattering elements are If the k2 surface element is visible and is not blocked by other scattering surface elements in the middle, then the first reflection wave of the k2 surface element to the k1 surface element is visible. If the condition is not met, then the first reflection wave of the k2 surface element to the k1 surface element is not visible. Among them, it is judged whether other scattering surface elements are It can be seen that the ① judgment is adopted, and the ② judgment is adopted to judge whether the middle part will be blocked by other scattering surface elements. They represent the unit normal vectors of the kth, k1th, and k2th scattering surface elements respectively, and LoS represents the unit observation vector of the radar perspective. is the unit vector of the line connecting the scattering centers of the two scattering planes, i is the unit direction vector of the incident wave, ζ is the preset detection threshold, r k′ is the normal vector of the scattering surface element that satisfies the occlusion condition, k b is the total number of scattering surface elements that meet the occlusion condition.

7. The accurate correction method for the flicker of electromagnetic scattering characteristics of ship images based on MIMO radar according to claim 6 is characterized by: The step 5 constructs a compensation dictionary based on the scattering surface elements obtained in step 4 that are visible to the incident electromagnetic wave and the primary reflected wave and are not blocked by other surface elements, and performs filtering processing on the radar echo signal, eliminates the echo signal of the primary reflected wave in the radar receiving signal, and compensates and corrects the echo detection deviation caused by the radar viewing angle and channel reciprocity, and then uses the sparse imaging algorithm to image the ship target to obtain the radar imaging result, including the following steps: Step 51: Use the scattering surface elements obtained in step 4 to build a compensation dictionary and filter the MIMO radar echo signal to construct a grid vector H; The scattering element obtained in step 4 is Scattering elements that are visible and not blocked by other scattering elements; First, obtain the MIMO radar echo: Among them, q is the path number of the first reflected wave, Q is the total number of paths of the first reflected wave, and A MP represents Q multipath reflections to the scattering surface element, is the path length of the qth primary reflection wave, is the real electromagnetic scattering intensity of K scattering surface elements, represents the RCS gain of the multipath signal on the scattering surface element k, is the potential artifact path, d0 is the real target echo, and d MP Indicates multipath echo, is a matrix of NLM×QK dimensions, represents the array flow matrix of the kth face element in the signal under the qth path, represents the array flow matrix of the multipath signal under the qth path, represents the scattering intensity of the kth bin in the multipath signal under the qth path, a 1 , b 1 、c 1 represents the steering vector of the first face element, a k , b k 、c k is the steering vector of the kth face element, a K , b K 、c K is the steering vector of the Kth face element, A is the array flow matrix of the real target signal, and M is the total number of transmitting array elements; Then, the scattering surface element obtained in step 4 is used to obtain the potential artifact path in the signal transmission and reception process. Thus construct the grid vector H; Step 52: Determine the weight of each element in H: Use the following formula to find the viewing direction of each scattering surface element: The corresponding artifact coordinates will be The weight at the corresponding artifact coordinate grid is set to 0, and the weight at other grids is set to 1; in, yes The corresponding artifact coordinates; Step 53: Use H after determining the weight of each element in step 52 and formula (13) to obtain the MIMO radar echo, and then construct the compensation dictionary A Comp ; Step 54: Process the echo received by the radar based on the compensation dictionary constructed in step 53, and convert formula (13) into the following formula: Step 5: Use formula (15) to construct a sparse reconstruction optimization problem and use the sparse imaging algorithm to obtain radar imaging results.

8. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 7 is characterized by: In step 53, the weight of each element is determined by using H and formula (13) to obtain the MIMO radar echo, and then construct the compensation dictionary A Comp , as follows: Among them, U represents the total number of grids divided in the three-dimensional scene, and Φ is the scattering deviation coefficient of each grid point in each observation channel, Represents the observation angle of the equivalent virtual channel with index m,n,l to the u-th grid.

9. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 8 is characterized by: The step 6, based on the radar imaging result obtained in step 5, further eliminates the defocus or distortion of the electromagnetic scattering characteristics of the scattering surface element caused by the movement of the ship, including the following steps: Step 61: Obtain the Doppler frequency shift in the MIMO radar echo signal: Where V = (v x ,v y ,v z ),ω=(ω x ,ω y ,ω z ) represent the three-dimensional linear velocity and angular velocity of the ship target, Represents the three-dimensional coordinates of the scattering center of the scattering surface element, V d It is the composite speed of the target's motion speed in each dimension; Step 62: Obtain the scattering surface element position offset ΔR caused by the ship motion on the radar echo signal phase according to the Doppler frequency shift obtained in step 61: f d t=2V d t / λ=2ΔR / λ (18) Among them, ΔR represents the three-dimensional coordinate offset of the face element Δx, Δy, Δz; Step 63: Get the deflection angle of the ship on the sea surface: i i (t)=A i sin(ω i t+c i ),i=roll,pitch,yaw (19) Among them, roll is the rolling mode, pitch is the pitching mode, yaw is the heading mode, i is the rotation mode, A i , γ i ,ω i denote the amplitude, initial phase and angular velocity of the deflection respectively; Step 64: Use the deflection angle of the ship on the sea surface obtained in step 63 to obtain the X, Y, Z three-dimensional coordinate system rotation matrix Ω of the ship's Roll, Pitch, and Yaw motions. r (t),Ω p (t),Ω y (t): Among them, r is the abbreviation of roll, p is the abbreviation of pitch, and y is the abbreviation of yaw; Step 65: Ω obtained according to step 64 r (t),Ω p (t),Ω y (t) Get the coordinate offset of the scattering surface element: Step 66: Eliminate the electromagnetic scattering characteristic distortion of the scattering surface element according to the coordinate offset obtained in step 65.

10. The accurate correction method for electromagnetic scattering characteristic flicker of ship image based on MIMO radar according to claim 9 is characterized by: The step sixty-six of eliminating the electromagnetic scattering characteristic distortion of the scattering surface element according to the coordinate offset obtained in step sixty-five is specifically as follows: First, using (19) to (21), we can obtain the ship's deflection angle at each sampling moment; Then, the three-dimensional position deflection vector Δr of the scattering center of the scattering surface element is obtained. k =(Δx k ,Δy k ,Δz k ), and according to Δr k The positive or negative value of LOS determines the deflection direction of each scattering surface element; Among them, Δr k LOS is positive, the deflection direction is toward the radar platform, Δr k LOS is negative, and the deflection direction is away from the radar platform; Finally, the electromagnetic scattering characteristics of each scattering surface element are compensated according to the deflection angle and deflection direction using formula (6), thereby eliminating the distortion of the electromagnetic scattering characteristics of the scattering surface element.

Images