Radar foresight super-resolution imaging method based on echo energy correction

Abstract

The invention discloses a radar foresight super-resolution imaging method based on echo energy correction, belongs to the technical field of radar imaging, and aims to solve the problem of imaging errors caused by speed errors of a traditional imaging method on a high-speed motion platform. The speed of the motion platform is accurately estimated based on an echo energy correction criterion, so that the influence of the motion error of the platform on the super-resolution imaging performance is effectively reduced, and the dependence of a traditional method on high-precision external speed measurement equipment is overcome. The method comprises the following steps: firstly, constructing a reconvolution model containing Doppler phase information; secondly, estimating the speed of the platform based on an echo energy correction criterion so as to obtain accurate speed information; then, updating a Doppler phase matrix by using the corrected speed, and correcting a reconvolution echo model; and finally, proposing a sparse super-resolution algorithm based on power weighting, and optimizing the target power and the noise power through alternate iteration to realize robust super-resolution imaging under the motion platform, thereby remarkably improving the target recognition and situation awareness capabilities of the motion platform.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of radar imaging, and particularly relates to a radar forward-looking super-resolution imaging method based on echo energy correction. BACKGROUND

[0002] In actual scenes such as aircraft landing in bad weather, ground mapping and target reconnaissance, forward-looking super-resolution imaging technology is crucial for target identification of a moving platform. However, due to the limitation of the real aperture size of a radar, the angular resolution of a traditional real beam radar is difficult to meet the demand. A deconvolution method converts the imaging problem into an inversion problem by modeling the echo as a convolution of the antenna pattern amplitude information and the target scattering coefficient, so as to improve the angular resolution of the forward-looking area.

[0003] Under the condition that the amplitude information of the antenna and the echo is known, researchers have proposed a variety of deconvolution methods based on different criteria. Zhang Q et al. proposed a sparse-based fast maximization minimization super-resolution algorithm, which, compared with the standard sparse maximization minimization method, improves the angular resolution of radar forward-looking imaging while speeding up the convergence.

[0004] The traditional convolution model reconstructs the target based on the antenna pattern amplitude information, but under a high-speed platform, the Doppler phase caused by platform motion will change the convolution process, thereby affecting the convolution matrix, leading to distortion of the imaging result or even false targets. Zhang H H et al. proposed a Bayesian forward-looking super-resolution imaging algorithm based on extended beam space Doppler deconvolution. This method combines the Doppler phase information of a high-speed moving platform with the antenna pattern characteristics, achieving efficient Doppler deconvolution. In the high-speed platform scenario, this method significantly improves the angular resolution and effectively suppresses noise. Y. Wu et al. proposed a complex deconvolution method that simultaneously uses amplitude and Doppler phase information. This method uses the correspondence between amplitude and Doppler phase to construct a complex convolution matrix, and uses the truncated singular value decomposition method to suppress noise amplification, achieving forward-looking super-resolution imaging of a moving platform.

[0005] These methods can achieve better imaging results under ideal platform motion states. However, in actual applications, platform speed is often affected by factors such as weather changes and platform acceleration, resulting in measurement errors. Currently, most radar forward-looking platforms rely on inertial navigation systems to measure flight speed to meet imaging needs, but high-precision measurement often requires high equipment costs. Therefore, it is particularly necessary to determine the platform motion speed through estimation methods. SUMMARY

[0006] To solve the above technical problems, the application provides a radar forward-looking super-resolution imaging method based on echo energy correction.

[0007] The specific technical scheme of the application is as follows:

[0008] 1. A radar forward-looking super-resolution imaging method based on echo energy correction, characterized in that the method comprises the following steps:

[0009] Step 1: the radar transmits a linear frequency modulation signal to a target area, and performs pulse compression and range migration correction on the received echo signal to obtain an echo signal model;

[0010] Step 2: a velocity estimation method based on echo energy correction, setting a velocity search interval, calculating a one-dimensional waveform entropy of the velocity search interval based on the obtained echo signal model, and iteratively converging the velocity search interval based on the one-dimensional waveform entropy until a convergence criterion is met to obtain a final estimated velocity;

[0011] Step 3: correcting the Doppler phase matrix of the echo signal based on the echo energy correction of the velocity, and updating the echo signal model;

[0012] Step 4: a sparse super-resolution algorithm based on power weighting, constructing a sparse constraint optimization problem based on power weighting according to the updated echo signal model and the target power matrix, iteratively solving, and obtaining a super-resolution imaging result.

[0013] The echo signal in the same distance unit r is formulated as follows:

[0014] ;

[0015] Wherein

[0016] ,

[0017] ;

[0018] In the formula, is a noise vector, is an echo signal representation, and is a fixed distance unit when the echo signal is collected at slow time t, represents the echo signal collected at the th slow time when the fixed distance unit is , wherein is the radar-target distance, is the round-trip time of the corresponding radar electromagnetic wave signal, denotes the position of the imaging point in the imaging scene in the ground coordinate system, and the superscript T denotes the transpose; is the radar-target distance, is the scattering coefficient of the different slow-time imaging points, represents the position of the m-th scattering point on the distance axis, denotes the angular dimension; denotes the m-th distance unit corresponding complex convolution matrix, which is constructed as follows:

[0019]

[0020] wherein, is a conventional convolution matrix constructed based on the antenna pattern, denotes the dot product operation, is a Doppler phase matrix constructed by the Doppler phase of the discretized echo signal.

[0021] The step 2 is specifically as follows:

[0022] Step 2.1: obtaining echo data which is composed of N columns of vectors obtained by N azimuth dimension sampling, and the dimension size is N, wherein, denotes the distance dimension sampling unit, denotes the azimuth dimension sampling unit; Step 2.2: setting the velocity search interval wherein,

[0023] is the lower limit of the velocity, is the upper limit of the velocity, which is set to , and the iteration termination threshold wherein, is the velocity measured according to the inertial navigation system; Step 2.3: respectively calculating the one-dimensional waveform entropy of the results of the velocity and

[0024] distance migration correction:

[0025] ;

[0026] ;

[0027] wherein, ,​​​​ for the velocity with the corrected one-dimensional waveform entropy, denotes a one-dimensional waveform entropy function, denotes an inverse fast Fourier transform, denotes a fast Fourier transform, j denotes the imaginary unit, denotes the radar's transmit signal frequency, denotes the squint angle of the radar beam center line of sight with respect to the boresight direction, is the speed of light;

[0028] Step 2.4: If , then shrink the search interval and then:

[0029] If , then shrink the interval as follows:

[0030] ;

[0031] and update the calculation ; where denotes the shrinkage ratio;

[0032] If , then shrink the interval as follows:

[0033] ;

[0034] and update the calculation ; iterate until the convergence criterion is satisfied;

[0035] Step 2.5: output the estimated velocity :

[0036] If at the termination of the iteration, , then ; otherwise, .

[0037] The one-dimensional waveform entropy function is specified as follows:

[0038]

[0039]

[0040] where is the normalized probability density of the signal, denotes the norm, G the discrete signal, and , is the length of the signal, The first vector of the discrete signal G is represented by A signal.

[0041] The Doppler phase matrix is ​​used to perform estimated velocity correction, and the formula is as follows:

[0042]

[0043] in is the corrected Doppler phase matrix; is the wavelength of the electromagnetic wave emitted by the radar, is the pitch angle, is the initial slant range of the target, For the 1st, ..., A slow time sampling moment, for The azimuth oblique angle corresponding to the moment; , the echo signal model is updated as:

[0044]

[0045] Where y is the echo signal representation and s is the scattering coefficient.

[0046] The step 4 is specifically as follows:

[0047] According to the updated echo signal model, let

[0048] ;

[0049] Where, Indicates echo signal The autocorrelation matrix of It is the echo signal The conjugate transpose of is a matrix The conjugate transpose of is a matrix No. column vectors, means to diagonalize the vector into a diagonal matrix, represents the target power matrix, Represents the diagonal vector of the target power matrix No. elements; then construct the following sparse constraint criteria:

[0050] ;

[0051] in represents the weight factor, The target The iteration loop is then minimized to achieve monotonic decrease and global convergence, and the final super-resolution imaging result is obtained:

[0052] ;

[0053] wherein represents the power of the target being updated, represents the power of the target being updated, represents the estimated value of the power of the target in the i th iteration, represents the conjugate transpose of represents the estimated echo covariance matrix in the i th iteration. represents the conjugate transpose of represents the estimated echo covariance matrix in the i th iteration. represents the estimated echo covariance matrix in the i th iteration.

[0054] The beneficial effects of the present application are as follows:

[0055] By using the echo energy correction criterion-based motion platform speed estimation, the influence of platform motion error on super-resolution imaging performance is effectively reduced, and the dependence on high-precision external speed measurement equipment in the traditional method is overcome. At the same time, the power-weighted sparse super-resolution algorithm is combined to iteratively optimize the target power and noise power, which significantly improves the imaging resolution and noise resistance performance. The method can obtain more stable, clear and accurate radar forward-looking super-resolution images in complex practical application scenarios, thereby significantly improving the target recognition and situation awareness capability of the motion platform. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 The flowchart of the present application.

[0057] Figure 2 The echo generation process analysis diagram of the present application.

[0058] Figure 3 The waveform entropy curve diagram of the present application.

[0059] Figure 4 The speed estimation error diagram under different signal-to-noise ratios of the present application.

[0060] Figure 5 The super-resolution result comparison diagram of different methods of the present application. DETAILED DESCRIPTION

[0061] The present application uses simulation experiments to demonstrate the effectiveness of the proposed method. All steps and conclusions of the present application are verified on the Matlab 2022b simulation platform. In order to enable relevant personnel in the field to understand the invention content, the present application is further described below in conjunction with the drawings.

[0062] ​A radar forward-looking super-resolution imaging method based on echo energy correction, as shown in Figure 1 comprises the following steps:

[0063] Step one, platform speed induced model error characterization method;

[0064] The radar transmits a linear frequency modulation signal to the target area, and pulse compression and range migration correction are performed on the received echo signal. After discretization processing, the azimuth echo signal can be represented as:

[0065]

[0066] wherein, is a double summation operator, the imaging scene is discretized into a plurality of scattering points on a two-dimensional grid, and the operator accumulates the scattering echo signals, represents the scattering coefficient of the target located at in the two-dimensional scene, is the position coordinate of the scattering point irradiated by the radar at the moment of the ground two-dimensional coordinate system, that is, the accurate time point at which the radar antenna beam center line of sight is aligned with the ground coordinate . represents fast time, represents slow time. The function represents antenna pattern modulation, represents carrier frequency. represents the function , is the bandwidth of the transmitted signal. is the initial slant range of the target, is the speed of light. is the instantaneous slant range between the radar antenna and the scattering point at the moment of slow time . represents the baseband signal obtained by the radar receiver.

[0067] The conventional deconvolution method usually ignores the Doppler phase , and the echo signal is constructed as the convolution of the antenna pattern amplitude information and the target scattering coefficient. However, on a high-speed platform, the Doppler phase affects the antenna modulation function, converts the convolution operation into a vector operation, causes false targets and reduces the angular resolution. To overcome this problem, the present application constructs a complex convolution matrix containing the Doppler phase.

[0068] Within the same range unit, the echo can be represented as:

[0069]

[0070] wherein

[0071]

[0072]

[0073] wherein represents a fixed distance unit as the first slow time collected echo signal, wherein is the distance between radar and target, and is the corresponding radar electromagnetic wave signal round trip time when represents the position of the imaging point in the ground coordinate system in the corresponding imaging scene, represents the scattering coefficient of different slow time imaging points when the distance between radar and target is represents the position of the M'th scattering point on the distance axis. M' represents the angular dimension, represents the first distance unit corresponding to the complex convolution matrix, which is constructed as follows:

[0074]

[0075] wherein, is the Doppler phase matrix constructed by the Doppler phase of formula (1). Symbol represents the dot product operation, is a conventional convolution matrix constructed based on the antenna pattern. Wherein the Doppler phase matrix is in the following form:

[0076]

[0077] wherein, is the imaginary unit, is the wavelength of the electromagnetic wave emitted by the radar, is the pitch angle, is the motion speed of the radar platform, is the first, …, slow time sampling moment, is the corresponding azimuth squint angle at moment.

[0078] Finally, the echo signal can be represented as:

[0079] ​

[0080] where, is the noise vector.

[0081] Step two, the velocity estimation method based on echo energy correction, the accurate estimation of velocity is realized by minimizing the waveform entropy. In super-resolution imaging, velocity error significantly affects the reconstruction performance of the target scene. The complex convolution model is based on the phase and amplitude information of the echo signal, while the velocity error will introduce phase distortion, destroy the accuracy of the model, cause imaging blur, and reduce the imaging quality. Therefore, in order to ensure the accuracy of super-resolution imaging, the platform velocity must be accurately estimated and compensated.

[0082] First, define the waveform entropy, for discrete signal , the waveform entropy is defined as:

[0083]

[0084] where, represents the th signal of the discrete signal G vector, , where is the normalized probability density of the signal, represents norm, G is the time sequence of the echo signal, is the length of the signal.

[0085] Waveform entropy can effectively evaluate the influence of motion parameters on the target distance direction. When the motion of the platform causes the peak value in the one-dimensional range image to diverge, the waveform entropy increases; when the velocity compensation error is zero, the waveform entropy reaches the minimum value. Therefore, by searching for the global minimum value of the waveform entropy on the velocity axis, the motion parameters of the platform can be determined.

[0086] In actual situations, due to the influence of noise, the entropy curve may appear pseudo-peak, affecting the performance of the interval search method. In order to solve this problem, the following formula optimizes the definition of the waveform entropy:

[0087]

[0088] where represents the optimized waveform entropy definition, which is distinguished from formula (6).

[0089] The specific steps of the velocity estimation method based on echo energy correction are as follows:

[0090] Step 2.1. Obtain the echo data after pulse compression , that is, obtained by azimuth dimension sampling ​Column vector composition, whose dimension size is wherein denotes the distance dimension sampling unit, denotes the azimuth dimension sampling unit.

[0091] Step 2.2. Set the velocity search interval wherein is the lower limit of velocity, is the upper limit of velocity, set as , and the iteration termination threshold wherein is the velocity measured according to the inertial navigation system, which has limited measurement accuracy and certain measurement error.

[0092] Step 2.3. Calculate the velocity and the velocity respectively.

[0093]

[0094]

[0095] wherein, , is the corrected one-dimensional waveform entropy of the velocity and denotes the one-dimensional waveform entropy function, which is defined as formula (7), i.e. , denotes the inverse fast Fourier transform, denotes the fast Fourier transform, denotes the radar transmitting signal frequency, denotes the squint angle of the radar beam center line of sight relative to the positive side view direction.

[0096] Step 2.4. If , then shrink the search interval;

[0097] If , then shrink the interval as follows:

[0098]

[0099]

[0100] If , then shrink the interval as follows:

[0101] ​

[0102]

[0103] in Represents the shrinkage ratio; after several iterations, until the convergence criterion is met , the speed search is terminated.

[0104] Step 2.5. Output the estimated speed :If the iteration is terminated, ,but On the contrary, .

[0105] Step 3: Correction of the complex convolution echo model;

[0106] After correcting the speed based on the echo energy, the present invention can obtain a more accurate platform motion speed than the inertial navigation measurement. , so a more accurate phase measurement matrix can be constructed. Furthermore, the Doppler phase matrix D can be modified as follows:

[0107]

[0108] in is the corrected Doppler phase matrix.

[0109] In order to simplify the modified echo model formula, let , so the echo signal can be updated as:

[0110]

[0111] Step 4: Sparse super-resolution algorithm based on power weighting;

[0112] According to the signal echo model shown in formula (15), let

[0113]

[0114] Where, Indicates echo signal The autocorrelation matrix of yes The conjugate transpose of yes The conjugate transpose of is a matrix No. column vectors, means to diagonalize the vector into a diagonal matrix, represents the target power matrix, Represents the diagonal vector of the target power matrix No. elements. Then construct the following sparse constraint criteria:

[0115]

[0116] in represents the weight factor, The target Power.

[0117] Equation (17) is a globally solvable convex optimization problem. Therefore, it can be minimized using iterative loops to achieve monotonically decreasing and global convergence in the iterations.

[0118]

[0119] No. The target power of the iteration can be expressed as:

[0120]

[0121] in Indicates that the target is being updated. Power, Indicates the In the first iteration, An estimate of the power, express The conjugate transpose of Indicates the At the iteration, the echo covariance matrix is ​​estimated.

[0122] Step 4: After iterative solution, the target super-resolution imaging result is obtained.

[0123] In order to verify the effectiveness of the present invention, simulation verification was carried out on the Matlab2022b simulation platform. The simulation scene of two-dimensional multi-point targets is as follows: Figure 2 As shown in (a) in the figure, the relevant simulation parameters are detailed in Table 1, and the simulation environment and platform are consistent with those described in Table 2. Figure 2 (b) in FIG shows the original echo signal, which covers multiple range cells. For the echo profile of a specific range cell, the present invention integrates the data along the azimuth direction to observe and verify the waveform entropy characteristics of the range signal. Figure 2 (c) in the figure is the echo image without range migration correction; Figure 2(d) in FIG. 6 is an echo image after distance migration correction; Figure 2 (e) and Figure 2 (f) in FIG. 6 are integrated sections after pulse compression without distance migration correction and after pulse compression with distance migration correction, respectively.

[0124] Table 1 Simulation parameters

[0125] Simulation parameters Value Carrier frequency 10.75 GHz Bandwidth 80 MHz Signal pulse width 2 μs Pulse repetition frequency 1000 Hz Antenna scanning speed 60° / s° Beam width 4° Scanning range -15°~15° Measurement platform speed 200 m / s

[0126] Table 2 Simulation environment

[0127] Hardware / software Parameter value CPU Inter i7-9700K RAM 64 G Simulation software Matlab 2022a

[0128] Based on the above analysis, different distance waveform entropy curves are obtained by changing the velocity compensation parameter as shown in FIG. 7, when the waveform entropy reaches the minimum value, the estimated velocity is very close to the actual velocity of the platform. Figure 3 Table 1 Simulation parameters Figure 4 It is shown that under the condition of low signal-to-noise ratio SNR=5 dB, the velocity estimation error is controlled within about 0.18 m / s by the method, thereby fully verifying the effectiveness of the velocity estimation method based on echo energy correction.

[0129] The application further adopts a sparse super-resolution algorithm based on power weighting to perform super-resolution reconstruction on the target scattering coefficient. A two-dimensional multi-point target simulation scene is shown in (a) of FIG. 8. The simulation parameters are shown in Table 3, and the simulation environment is consistent with Table 2. Figure 5

[0130] Table 3 Two-dimensional multi-point target simulation parameters

[0131] Simulation parameters Value Carrier frequency 10.75 GHz Bandwidth 80 MHz Signal pulse width 2 μs Pulse repetition frequency 1000 Hz Antenna scanning speed 60° / s° Beam width 4° Scanning range -15°~15° Platform speed 200 m / s Measurement speed error 0.5 m / s

[0132] The targets are distributed as follows within four distance units: one target is located at 0° and has a distance of 3.84 km; two targets are respectively located at -0.5° and 0.5° and have a distance of 3.72 km; four targets are respectively located at -1.5°, -0.5°, 0.5° and 1.5° and have a distance of 3.6 km; six targets are respectively located at -2.5°, -1.5°, -0.5°, 0.5°, 1.5° and 2.5° and have a distance of 3.48 km. Figure 5 (b) shows the real beam result with a signal-to-noise ratio of 10 dB, and after pulse compression and distance offset correction, the targets in the same distance unit are difficult to distinguish. Figure 5 (c) in FIG. 8 is an iterative adaptive algorithm (IAA) method, and false targets exist in the imaging result. Figure 5 (d) and Figure 5 ​(e) in the table is sparse iterative covariance estimation (SPICE) and sparse learning iterative minimization (SLIM) method respectively, the super-resolution effect is better, but the performance improvement is limited, and false targets exist. Figure 5 (f) in the table is the result of the proposed method, which can accurately restore the target and effectively suppress noise.

[0133] In addition, the effectiveness of the proposed method is further verified by two indicators of mean square error (MSE) and structural similarity (SSIM). The evaluation index values of different methods are shown in Table 4.

[0134] MSE is defined as:

[0135]

[0136] In the formula, and denote the sampling points of distance and azimuth, denote the number of Monte Carlo experiments. and denote the estimated value and the true target scattering coefficient of the target scattering coefficient respectively.

[0137] SSIM is defined as:

[0138]

[0139] In the formula, and are the mean and standard deviation of the vector sum respectively, is the correlation coefficient of the vector and .

[0140] As can be seen from Table 4, the MSE of the proposed method is significantly lower than that of other methods, and the SSIM is significantly improved, verifying the estimation accuracy and effectiveness of the method.

[0141] Table 4: Mean square error and structural similarity index of imaging results of different methods

[0142] Method MSE (dB) SSIM Real beam -37.0253 0.0591 IAA -63.6191 0.9697 SPICE -67.3389 0.9712 SLIM -68.0491 0.9737 Proposed method -74.5322 0.9901

[0143] The skilled engineer can make related applications according to the disclosed radar forward-looking super-resolution imaging method based on echo energy correction, and the related knowledge is still within the protection scope of the present application.

[0144] It is to be understood that the present application is described by way of example only, and that modifications or alterations can be made to the features and embodiments described without departing from the spirit and scope of the application. In addition, modifications can be made to the features and embodiments described to accommodate specific situations and materials without departing from the spirit and scope of the application. Accordingly, the application is not limited to the specific embodiments disclosed herein, but rather, the scope of the application includes all embodiments falling within the scope of the claims.

Claims

1. A radar forward-looking super-resolution imaging method based on echo energy correction, characterized in that: The following steps are involved: Step 1: The radar transmits a linear frequency modulation signal to the target area and performs pulse compression and range migration correction on the received echo signal to obtain the echo signal model; Step 2: Based on the velocity estimation method of echo energy correction, a velocity search interval is set. Based on the obtained echo signal model, the one-dimensional waveform entropy of the velocity search interval is calculated. The velocity search interval is iteratively converged based on the one-dimensional waveform entropy until the convergence criterion is met to obtain the final estimated velocity. Step 3: Correct the Doppler phase matrix of the echo signal based on the velocity of the echo energy correction and update the echo signal model; Step 4: Based on the power-weighted sparse super-resolution algorithm, a power-weighted sparse constrained optimization problem is constructed according to the updated echo signal model and the target power matrix, and the problem is solved iteratively to obtain the super-resolution imaging result.

2. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 1, characterized in that: The echo signal is within the same distance unit r, and the echo signal model is formulated as follows: ； in ， ； Where, is the noise vector, is the echo signal representation, and is the fixed distance unit. The echo signal collected at time t is Indicates that the fixed distance unit is Time Slow time The collected echo signal, The distance between the radar and the target is When , the corresponding round trip time of the radar electromagnetic wave signal is, represents the position of the imaging point in the corresponding imaging scene in the ground coordinate system, and the superscript T represents transposition; Indicates the distance between the radar and the target is The scattering coefficients of different slow time imaging points are: Represents the distance to the axis The scattering point locations, represents the angular dimension; Indicates the The complex convolution matrix corresponding to the distance unit is constructed as follows: ； in, is a conventional convolution matrix constructed based on the antenna pattern, represents the dot product operation, is the Doppler phase matrix, which is constructed from the Doppler phase of the discretized echo signal.

3. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 2, characterized in that: The step 2 is specifically as follows: Step 2.1: Acquire echo data , which is due to The azimuth dimension sampling Column vectors, dimension size is ,in represents the distance dimension sampling unit, represents the azimuth dimension sampling unit; Step 2.2: Set the speed search range ,in is the lower speed limit, The upper speed limit is set to , , and the iteration termination threshold ,in is the speed measured by the inertial navigation system; Step 2.3: Calculate the speed separately and The one-dimensional waveform entropy of the result of range migration correction is: ； ； in, 、 For speed and The corrected one-dimensional waveform entropy, represents the one-dimensional waveform entropy function, represents the inverse fast Fourier transform, represents the fast Fourier transform, j represents the imaginary unit, Indicates the frequency of the radar's transmitted signal, It indicates the oblique angle of the radar beam center line of sight relative to the positive side view direction. is the speed of light; Step 2.4: If , then shrink the search interval, and then: like , the contraction interval is as follows: ； And update the calculation ;in Indicates the shrinkage ratio; like , the contraction interval is as follows: ； And update the calculation ; After iteration, until the convergence criterion is met , then terminate the speed search; Step 2.5: Output the estimated speed : If the iteration is terminated, ,but On the contrary, .

4. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 3, characterized in that: The one-dimensional waveform entropy function is specifically as follows: ； ； in, is the normalized probability density of the signal, express norm, G is a discrete signal, and , is the length of the signal, The first vector of the discrete signal G is represented by A signal.

5. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 4, characterized in that: The Doppler phase matrix is ​​used to perform estimated velocity correction, and the formula is as follows: ； in is the corrected Doppler phase matrix; is the wavelength of the electromagnetic wave emitted by the radar, is the pitch angle, is the initial slant range of the target, For the 1st, ..., A slow time sampling moment, for The azimuth oblique angle corresponding to the moment; , the echo signal model is updated as: ； Where y is the echo signal representation and s is the scattering coefficient.

6. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 5, characterized in that: The step 4 is specifically as follows: According to the updated echo signal model, let ； Where, Indicates echo signal The autocorrelation matrix of It is the echo signal The conjugate transpose of is a matrix The conjugate transpose of is a matrix No. column vectors, means to diagonalize the vector into a diagonal matrix, represents the target power matrix, Represents the diagonal vector of the target power matrix No. elements; then construct the following sparse constraint criteria: ； in represents the weight factor, The target power; then the iterative cycle is minimized to achieve monotonically decreasing and global convergence, and the final super-resolution imaging result is obtained: ； in Indicates that the target is being updated. Power, Indicates the In the first iteration, An estimate of the power, express The conjugate transpose of Indicates the At the iteration, the echo covariance matrix is ​​estimated.

7. The radar forward-looking super-resolution imaging method based on echo energy correction according to claim 6, characterized in that: No. The target power of the iteration can be expressed as: 。

Images