Dual-base ISAR back-projection imaging method based on image self-focusing

Abstract

The application discloses a double-base ISAR back-projection imaging method based on image self-focusing, relates to the technical field of radar signal processing, and solves the problems that the envelope error and the phase error caused by the translational motion error are not considered in the prior art and the error in BP imaging cannot be compensated. The method comprises the following steps: firstly, the translational motion is estimated and compensated, so that the defocusing influence of the translational motion on the imaging result is reduced; then, the image sharpening degree is used as an evaluation index, an optimization function is established, the error phases of all pulses are optimized, and the phase error is compensated; since the coordinate descent method is adopted, the closed solution of the pulse phase error can be obtained each time, after the phase errors of all pulses are solved, the envelope error and the phase error can be further uniformly compensated, and the accurate optimization solution can be obtained after a small number of iterations; and the method realizes the compensation of the envelope error and the phase error caused by the translational motion error in the target echo.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and in particular to a bistatic ISAR back projection imaging method based on image autofocus. Background Technology

[0002] Due to its all-weather, all-day, high-resolution, and long-range capabilities, Inverse Synthetic Aperture Radar (ISAR) plays a crucial role in airborne and space target detection. With the rapid development of ISAR, although existing imaging radars offer high resolution, even higher resolution is needed to accurately describe the features of small satellites and other space targets. Backprojection is a precise temporal imaging algorithm that has played a significant role in cooperative SAR imaging. However, bistatic ISAR imaging relies on the relative motion between the target and the two radars, and the non-cooperative nature of the target introduces many undesirable factors into the imaging process. Particularly for moving targets, complex motion compensation makes bistatic ISAR imaging difficult, causing the compensation phase factor to fail, increasing the error of the backprojection algorithm, and ultimately resulting in defocused images and unsatisfactory imaging results.

[0003] Dong Huixu et al.'s "MIMO-ISAR Two-Dimensional Imaging Method Based on Linear Array" discloses an imaging method that uses lateral focusing of data from the same range cell. This method avoids interpolation and equivalent array element error compensation calculations in imaging and achieves lateral calibration of the target image. However, this method does not consider the accuracy of translational compensation and phase compensation, and does not process the errors generated by compensation, resulting in low imaging quality. Li Yi's "Fast Backprojection Algorithm for MIMO-ISAR Based on FFT" discloses a fast backprojection algorithm for MIMO-ISAR based on FFT, providing constraints for converting nonlinear phase compensation factors into linear compensation factors, which greatly reduces the computation time of the MIMO-ISAR backprojection algorithm and improves its real-time performance. However, even under the constraints, this algorithm still has conversion errors, resulting in low imaging quality; beyond the constraints, the conversion error increases, and the sidelobe level in the imaging will rise significantly.

[0004] The shortcomings of existing technologies are: 1. After translational compensation, directly using the BP algorithm to perform bistatic ISAR imaging of the target does not consider the influence of phase error caused by translational error. 2. When performing BP imaging for bistatic ISAR, existing technologies cannot effectively compensate for envelope error and phase error in a unified manner. Summary of the Invention

[0005] This invention provides a bistatic ISAR back projection imaging method based on image autofocus, which solves the problems in the prior art that do not consider the phase error caused by translational error and cannot compensate for the error in BP imaging, and realizes the compensation of envelope error and phase error caused by translational error in the target echo.

[0006] This invention provides a method for bistatic ISAR backprojection imaging based on image autofocus, the method comprising:

[0007] The radar echo signal is acquired, and the radar echo signal is processed using line frequency modulation demodulation to obtain a bistatic ISAR echo signal matrix; wherein, the radar echo signal includes: multiple pulse echoes;

[0008] The phase term in the bistatic ISAR echo signal matrix is ​​analyzed to obtain the translational and rotational components, and the translational components are compensated to obtain the translational compensation matrix.

[0009] The dot product of the bistatic ISAR echo signal matrix and the translational compensation matrix is ​​performed to obtain the translationally compensated echo signal; wherein, the translationally compensated echo signal includes: translational error;

[0010] The Doppler center is estimated and corrected on the translational compensation echo signal to eliminate the linear term in the translational error and obtain the first corrected echo signal.

[0011] A BP imaging projection network is constructed based on the rotation component, and BP imaging is performed on the first corrected echo signal to obtain a BP imaging result; wherein, the BP imaging result is the sum of the back projection values ​​of multiple pulse echoes of the first corrected echo signal;

[0012] The evaluation index and optimization model are determined, the optimization model is iteratively solved using the coordinate descent method, and the solution results are evaluated using the evaluation index. The phase estimation set is obtained based on the evaluation results.

[0013] The first corrected echo signal is compensated according to the phase estimation set to obtain the corrected echo signal, and the corrected echo signal is subjected to BP imaging to obtain the final imaging result.

[0014] In one possible implementation, acquiring the radar echo signal and processing the radar echo signal using line frequency modulation demodulation to obtain a bistatic ISAR echo signal matrix includes:

[0015] The radar echo signal is compressed by range pulse compression using demodulation frequency modulation to obtain a compressed signal.

[0016] The RVP phase term and envelope skew phase term of the compressed signal are compensated to obtain the bi-base ISAR echo signal matrix.

[0017] In one possible implementation, the bistatic ISAR echo signal matrix is ​​represented as:

[0018]

[0019] in, The scattering coefficient of the scattering point in the direction of the receiving radar is represented by σ. i Phase is t m For slow time series, -M / 2 ≤ m ≤ M / 2-1; t k For fast time series, -N / 2 ≤ k ≤ N / 2-1; f c T is the center frequency of the transmitted signal; p γ is the pulse width of the transmitted signal; γ is the frequency modulation; ΔR i This represents the sum of the distances from the two radars to the scattering point and the sum of the distances from the two radars to the reference point.

[0020] In one possible implementation, the step of analyzing the phase terms in the bistatic ISAR echo signal matrix to obtain translational and rotational components, and compensating for the translational components to obtain a translational compensation matrix, includes:

[0021] The initial phase term expression in the bistatic ISAR echo signal matrix is ​​determined, and the linearity of the radar echo signal is used to modify and derive the initial phase term expression to obtain the derived expression.

[0022] The derived expression is decomposed to obtain the translational component and the rotational component;

[0023] The translational components are compensated using the translational compensation method to obtain the translational compensation matrix.

[0024] In one possible implementation, the translational error is expressed as:

[0025]

[0026] Where, μ(t) m ) for slow time t m The relevant polynomial; γ is the modulation frequency; f c The center frequency of the transmitted signal; t m For slow time series, -M / 2 ≤ m ≤ M / 2-1; t k For a fast time series, -N / 2≤k≤N / 2-1; c is the speed of light.

[0027] In one possible implementation, the step of constructing a BP imaging projection network based on the rotation component and performing BP imaging on the first corrected echo signal to obtain the BP imaging result includes:

[0028] Determine the rotation component of the echo signal at any scattering point, and construct a BP imaging projection grid based on the rotation component;

[0029] Perform a distance-dimensional Fourier transform on the first corrected echo signal to obtain the transform result; wherein the transform result is related to the rotation component;

[0030] The transformation result is then subjected to BP imaging on the BP imaging projection grid to obtain the BP imaging result.

[0031] In one possible implementation, the transformation result is specifically represented as follows:

[0032]

[0033] Among them, A i T represents the scattering coefficient of the scattering point in the direction of the receiving radar; p Indicates the pulse width of the transmitted signal; γ represents the frequency modulation rate; c represents the speed of light; f k The frequency difference represents the coherent difference frequency; r represents the sum of the rotational phase and the translational error phase; f represents the frequency difference frequency. c This indicates the center frequency of the transmitted signal.

[0034] In one possible implementation, the BP imaging result is expressed as:

[0035]

[0036] Where S2 represents the result after the distance-to-Fourier transform; f k The coherent difference frequency is represented by θ; the rotation angle of the target relative to the equivalent monostatic radar over a slow time interval is represented by f. c Indicates the carrier frequency; β represents the bipolar angle; t m For a slow time series, -M / 2≤m≤M / 2-1; x represents the X coordinate of the projected grid; y represents the Y coordinate of the projected grid.

[0037] In one possible implementation, the evaluation metric includes image sharpness, specifically expressed as:

[0038]

[0039] Among them, v η Let η be the image intensity of the nth pixel. This represents the phase estimation set; N represents the number of range units of the echo.

[0040] In one possible implementation, the step of compensating the first corrected echo signal according to the phase estimation set to obtain the corrected echo signal includes:

[0041] A phase estimation set is obtained by using a self-focusing algorithm based on the maximum image sharpness;

[0042] The phase compensation factor is obtained using the phase estimates from the aforementioned phase estimation set.

[0043] The first corrected echo signal is compensated according to the phase compensation factor to obtain the corrected echo signal.

[0044] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0045] (1) This invention utilizes the self-focusing method and the coordinate descent method to obtain the closed solution of each pulse. Therefore, the accuracy of each solution is high. By adopting the idea of ​​one-cycle iterative processing, the optimal solution of phase compensation for each pulse can be obtained in a few iterations, and the convergence stability is greatly enhanced.

[0046] (2) This invention takes into account the influence of translational error of bistatic ISAR on envelope error and phase error, uses image sharpness as evaluation index, establishes optimization function, optimizes the error phase of each pulse, and compensates for phase error. Attached Figure Description

[0047] Figure 1 A flowchart of the method steps for bistatic ISAR back projection imaging based on image autofocus provided in an embodiment of the present invention;

[0048] Figure 2 The simulation point scattering model provided in the embodiments of the present invention;

[0049] Figure 3 This is a schematic diagram of the motion geometry of the target in a simulation experiment provided in an embodiment of the present invention;

[0050] Figure 4 The direct BP imaging results provided in the embodiments of the present invention;

[0051] Figure 5 This is the final imaging result after self-focusing compensation provided in the embodiments of the present invention;

[0052] Figure 6 The positions of marker points A, B, and C provided in the embodiments of the present invention;

[0053] Figure 7 This is a contour map of point A after compensation (4 times interpolation) provided in an embodiment of the present invention.

[0054] Figure 8The contour map of point B after compensation (4 times interpolation) provided in this embodiment of the invention;

[0055] Figure 9 This is a contour map of the 4x interpolation point C after compensation, provided in an embodiment of the present invention.

[0056] Figure 10 The azimuth slice and distance slice of points A, B, and C provided in the embodiments of the present invention;

[0057] Figure 11 The image sharpness value iteration curve is provided by the autofocus algorithm based on maximum image sharpness in the embodiments of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] The present invention provides a method for image autofocus-based bistatic ISAR back projection imaging, as shown in the figure. The method includes the following steps S101 to S107.

[0060] S101, acquire the radar echo signal, process the radar echo signal using linear frequency modulation demodulation, and obtain the bistatic ISAR echo signal matrix; wherein, the radar echo signal includes: multiple pulse echoes. Here, in the bistatic inverse synthetic aperture radar (ISAR), after radar A transmits a linear frequency modulated signal to the moving target, radar B acquires the received signal matrix scattered by the target under low signal-to-noise ratio conditions, wherein the size of the echo matrix is ​​N×M, where N represents the number of range cells of the echo and M represents the number of azimuth cells of the echo.

[0061] Specifically, in step S101, the radar echo signal is acquired, and the radar echo signal is processed by demodulation to obtain the bistatic ISAR echo signal matrix, including the following steps S1011 to S1012.

[0062] S1011 uses demodulation frequency modulation to perform range pulse compression on the radar echo signal to obtain a compressed signal.

[0063] S1012 compensates for the RVP phase term and envelope skew phase term of the compressed signal to obtain the bi-base ISAR echo signal matrix.

[0064] For example, by using the de-frequency modulation method, range pulse compression processing is performed on the echo signal, and the RVP phase term and envelope skew phase term are compensated to obtain the bi-base ISAR echo signal matrix:

[0065]

[0066] in, The scattering coefficient of the scattering point in the direction of the receiving radar is represented by σ. i Phase is t m For slow time series, -M / 2 ≤ m ≤ M / 2-1; t k For fast time series, -N / 2 ≤ k ≤ N / 2-1; f c T is the center frequency of the transmitted signal; p γ is the pulse width of the transmitted signal; γ is the frequency modulation; ΔR i This represents the difference between the sum of the distances from the two radars to the scattering point and the sum of the distances from the two radars to the reference point, i.e.:

[0067]

[0068] Among them, R i (t m R represents the sum of the slant ranges from the scattering point to the two radars. a (t m ) and R b (t m ) represent the distances from the two radars to the origin of the target's coordinate system, and x represents the translational component. i =(x i ,y i ,z i () represents the coordinates of point i in a coordinate system with the center of the target as the origin, i a (t m ) and i b (t m ) represent the line-of-sight directions of the two radars, R ref This represents the sum of the distances between the reference point and the two radars.

[0069] S102, the phase term in the bistatic ISAR echo signal matrix is ​​analyzed to obtain the translational and rotational components, and the translational components are compensated to obtain the translational compensation matrix.

[0070] Specifically, in step S102, the phase term in the bistatic ISAR echo signal matrix is ​​analyzed to obtain the translational and rotational components, and the translational components are compensated to obtain the translational compensation matrix, including the following steps S1021 to S1023.

[0071] S1021, determine the initial phase term expression in the dual-base ISAR echo signal matrix, and derive the derived expression by modifying the initial phase term expression using the linear property of the radar echo signal. Here, the derived expression is specifically expressed as:

[0072]

[0073] Where γ is the modulation frequency; ΔR i The sum of the distances from the two radars to the scattering point and the sum of the distances from the two radars to the reference point represent the difference; f c The center frequency of the transmitted signal; t k is a fast time series; c is the speed of light.

[0074] S1022, decompose the derived expression to obtain the translational and rotational components. For example, since the echo signal received in continuous coherent reception has linear properties, the distance difference ΔR... i and the distance R from the scattering point i to the radar i (t m The difference is only one constant value R. ref Therefore, ΔR i Use R i (t m The substitution will not affect the overall azimuth imaging analysis results, that is:

[0075]

[0076] in,

[0077]

[0078] β(t m )=∠(i a (t m ),i b (t m ))=arccos(i a (t m )·i b (t m (6)

[0079]

[0080] Where β represents the bibasic angle, which is determined by t. m Changeable, i c (t m This indicates the radar line-of-sight direction of the equivalent monostatic ISAR; while The reference phase term, representing the position at the origin of the target's coordinate system, is a translational component; This indicates the coordinate position of the target caused by rotation at x.i (x i ,y i ,z i The phase of the scattering point i at () is the rotation component.

[0081] S1023, use the translational compensation method to compensate for the translational components to obtain the translational compensation matrix. For example, use the translational compensation method to compensate for the translational components and calculate each element in the translational compensation matrix.

[0082] S103, perform a dot product between the bistatic ISAR echo signal matrix and the translational compensation matrix to obtain the translationally compensated echo signal; wherein, the translationally compensated echo signal includes: translational error.

[0083] Specifically, the translational error is expressed as:

[0084]

[0085] Where, μ(t) m ) for slow time t m The relevant polynomial; γ is the modulation frequency; f c The center frequency of the transmitted signal; t m For slow time series, -M / 2 ≤ m ≤ M / 2-1; t k For a fast time series, -N / 2≤k≤N / 2-1; c is the speed of light.

[0086] μ(t m The polynomial related to the slow time tm can be expressed as:

[0087]

[0088] Among them, ξ1, ξ2, ξ3, and ξ4 are all constants.

[0089] S104 performs Doppler center estimation and correction on the translational compensation echo signal to eliminate the linear term in the translational error and obtain the first corrected echo signal.

[0090] This invention considers the phase error caused by translational error, uses a self-focusing method to construct an optimization function, solves for the error phase, and then compensates for the envelope error and phase error, significantly improving the final dual-base ISAR focusing quality.

[0091] S105, construct a BP imaging projection network based on the rotation component, perform BP imaging on the first corrected echo signal, and obtain the BP imaging result; wherein, the BP imaging result is the sum of the back projection values ​​of multiple pulse echoes of the first corrected echo signal.

[0092] Specifically, a BP imaging projection network is constructed based on the rotation component, and BP imaging is performed on the first corrected echo signal to obtain the BP imaging result, including the following steps S1051 to S1053.

[0093] S1051, determine the rotation component of the echo signal from any scattering point, and construct a BP imaging projection grid based on the rotation component. For example, the phase of the rotation component of the echo signal from any ideal scattering point i on the target. It can be represented as

[0094]

[0095] In the formula θ(t) m (x') represents the effective rotation angle of the target. i ,y' i Let be the projection of the coordinates of point i onto the imaging plane.

[0096] S1052, Perform a range-dimensional Fourier transform on the first corrected echo signal to obtain the transform result; wherein the transform result is related to the rotation component. For example, according to equation (1), perform a range-dimensional Fourier transform on the first corrected echo signal, i.e.:

[0097]

[0098] in,

[0099]

[0100]

[0101] Among them, A i T represents the scattering coefficient of the scattering point in the direction of the receiving radar; p Indicates the pulse width of the transmitted signal; γ represents the frequency modulation rate; c represents the speed of light; f k The frequency difference represents the coherent difference frequency; r represents the sum of the rotational phase and the translational error phase; f represents the frequency difference frequency. c This indicates the center frequency of the transmitted signal.

[0102] S1053, perform BP imaging on the BP imaging projection grid to obtain the BP imaging result. For example, directly perform BP imaging on formula (13), i.e.:

[0103]

[0104] Where S2 represents the result after the distance-to-Fourier transform; f k The coherent difference frequency is represented by θ; the rotation angle of the target relative to the equivalent monostatic radar over a slow time interval is represented by f. c Indicates the carrier frequency; β represents the bipolar angle; t mFor a slow time series, -M / 2≤m≤M / 2-1; x represents the X coordinate of the projected grid; y represents the Y coordinate of the projected grid.

[0105] S106. Determine the evaluation index and optimization model, use the coordinate descent method to iteratively solve the optimization model, and use the evaluation index to evaluate the solution results. Based on the evaluation results, obtain the phase estimation set.

[0106] For example, based on the above analysis, due to the influence of translational errors, if BP imaging is performed directly, errors will occur in each echo phase term: This causes the image to go out of focus.

[0107] Phase estimation is required to focus the image. And solve for the phase compensation function: The imaging results are compensated to eliminate the effects of envelope and phase errors. Assume the back projection values ​​of the focused image are as follows:

[0108]

[0109] Where j = 1, 2, ..., M represents the number of each echo pulse, b j The backward projection result of the j-th pulse can be represented as:

[0110]

[0111] This invention provides a method to achieve step S106, wherein the goal of the self-focusing algorithm is to generate a phase estimate. Image sharpness is chosen as the evaluation metric. Image sharpness is defined as the square of the image intensity, and its expression is:

[0112]

[0113] Among them, v η The image intensity of the ηth pixel is calculated using the following formula:

[0114]

[0115] The optimization model is then expressed as:

[0116]

[0117] Since the optimization of equation (19) has no closed-form solution, the coordinate descent method is used for iterative solution. Let... Let represent the estimated value of the j-th pulse in the nth iteration. Then, the phase estimate for the (n+1)-th iteration is:

[0118]

[0119] In this case, the back projection value of the image is:

[0120]

[0121] Where x represents the sum of the backward projections of all pulses except the j-th pulse, and y represents the back projection value of the j-th pulse without motion error compensation. Then the intensity of the η-th pixel is...

[0122]

[0123] Since the first two terms are constants, we can define v0 and v φ Two vectors, where the elements are defined as:

[0124]

[0125] Therefore, equation (22) can be written as:

[0126] v = v0 + v φ (twenty four)

[0127] Where v0 is a constant, v φ It is related to the phase of the j-th pulse. Since the L2 norm can be expressed as a sum of squares in mathematics, the formula for calculating image sharpness can be expressed as:

[0128]

[0129] Among them, v η Let η be the image intensity of the nth pixel. This represents the phase estimation set; N represents the number of range units of the echo.

[0130] The L2 norm can represent the straight-line distance between two vector matrices in space, so the maximum image sharpness can be converted into maximizing the length of vector v.

[0131] To simplify v φ Define a = (a1, a2, ..., a2) N ) and b = (b1, b2, ..., b N Two vectors, where the elements of the vectors are:

[0132]

[0133] Then v φ Represented as:

[0134] v φ = a cosφ + b sinφ (27)

[0135] Formula (27) has a similar form to the ellipse, even though vectors a and b are not necessarily orthogonal. Let the plane formed by vectors a and b be the γ plane. Define a new two-dimensional coordinate system with the center of the ellipse as the origin in this plane. Vector v0 can be decomposed into components in the β plane and components perpendicular to the γ plane, with the foot of the perpendicular being x0. Since the perpendicular component is invariant, its influence is ignored. For easier description, define a set of unit orthogonal basis vectors {e1,e2} in the γ plane, and their specific expressions are as follows:

[0136]

[0137] Then the coordinates of vectors a and b under this basis can be expressed as:

[0138]

[0139] In the formula, E = [e1, e2], and the coordinates of x0 in the new coordinate system are x0 = -E. T v0.

[0140] Therefore, the optimization problem is transformed into finding a point on the ellipse spanned by vectors a and b. This maximizes the distance between it and point x0. The following formula should be satisfied:

[0141]

[0142] In the formula, For an ellipse at point The normal vector of , where α is a constant.

[0143] The parametric equation of an ellipse can be expressed as:

[0144]

[0145] Or it can be expressed as:

[0146] f(x) = x T Rx = 1 (32)

[0147] In the formula, matrix R can be represented as:

[0148]

[0149] The element values ​​are:

[0150]

[0151] Therefore, the normal vector of the ellipse can be expressed as:

[0152]

[0153] Substituting equation (35) into equation (30), and incorporating 2 into the unknown parameter α, we can obtain

[0154]

[0155] Perform eigenvalue decomposition on matrix R, R = VΛV T Λ = diag({λ1,λ2}), where V is a matrix composed of eigenvectors, and λ1 and λ2 are eigenvalues. Substituting the eigenvalue decomposition of matrix R into equation (36) and then into the ellipse equation, we define the vector [β1,β2]. T =v T x0, rearranging, we get:

[0156]

[0157] The parameters are as follows:

[0158]

[0159] Solving equation (37) yields the smallest real solution for α. Bundle Substituting the value into the formula, the phase error can be obtained:

[0160]

[0161] in, and The value is:

[0162]

[0163] The self-focusing method employed in this invention utilizes a coordinate descent approach to obtain the closed-loop solution for each pulse, resulting in high accuracy for each solution. By employing a single-loop iterative process, the optimal phase compensation solution for each pulse can be obtained with only a few iterations. This significantly enhances convergence stability.

[0164] S107, the BP imaging results are compensated according to the phase estimation set to obtain the corrected echo signal, and the corrected echo signal is then subjected to BP imaging to obtain the final imaging result.

[0165] Specifically, in step S107, the first corrected echo signal is compensated according to the phase estimation set to obtain the corrected echo signal, including the following steps S1071 to S1073.

[0166] S1071 uses a self-focusing algorithm based on the maximum sharpness of the image to obtain a phase estimation set.

[0167] S1072, using the phase estimates in the phase estimate set, obtain the phase compensation factor.

[0168] S1073, the first corrected echo signal is compensated according to the phase compensation factor to obtain the corrected echo signal.

[0169] For example, the error polynomial vector corresponding to each pulse can be solved from the phase estimation set. Right now:

[0170]

[0171] Then construct phase The translational error is compensated to eliminate the envelope and phase errors it causes, and finally, BP imaging is performed again.

[0172] To increase the accuracy of the compensation, steps S105 to S107 can be repeated 3 to 4 times.

[0173] The following is an analysis of the simulation content and results provided by this invention.

[0174] The effectiveness of this invention can be further illustrated by the following simulation.

[0175] (11) Simulation content and result analysis:

[0176] Simulation Experiment: The simulated radar parameters are shown in Table 1, and the simulated two-dimensional point scattering model is as follows. Figure 2 As shown. The geometric model of the simulation scene is as follows. Figure 3 As shown, the target moves in a straight line along the X-axis with a velocity of 500 m / s and an acceleration of 400 m / s². 2 The receiving radar and the transmitting radar are both located along the X-axis, and the distance between them is 10 km. The equivalent single-base radar is located between the transmitting radar and the receiving radar.

[0177] Table 1 Simulation Radar Parameter Table

[0178] carrier frequency bandwidth Pulse repetition frequency Pulse width pulse count 12GHz 1GHz 800Hz 10us 512

[0179] like Figure 4 As shown, the target exhibits severe azimuth defocus in the direct BP imaging results. Figure 5 The image shows the processing result of the autofocus method based on maximum image sharpness. The imaging results demonstrate that this algorithm can achieve image focusing. Edge points A and B and center point C are obtained as follows: Figure 6 As shown, after interpolating by a factor of 4, the resulting contour map is as follows. Figure 7 As shown in Figures 8 and 9, it can be seen that points A, B, and C are all well focused.

[0180] To better analyze the focusing effect, Figure 10 The azimuth and range profiles of points A, B, and C are given. Figure 10(a), 10(c), and 10(e) are the azimuth profiles of points A, B, and C, respectively. The simulation results show that the azimuth profile of the image before focusing is severely broadened and the sidelobe energy is enhanced. After compensation using the proposed algorithm, the problems of azimuth broadening and sidelobe enhancement in the azimuth profile are significantly improved. Figure 10 (b), 10(d), and 10(f) are the distance profiles of points A, B, and C, respectively. The simulation results show that the distance profiles remain basically unchanged before and after focusing, indicating that the translational error has little effect on the distance direction and does not cause the image to defocus along the distance direction.

[0181] To quantify the focusing effect, Table 3 presents the Peak-Lobe Ratio (PSLR) and Integrated Side-Lobe Ratio (ISLR) in the azimuth direction before and after compensation for points A, B, and C, accurately and intuitively illustrating the effectiveness of the proposed algorithm. Table 4 presents the Peak-Lobe Ratio and Integrated Side-Lobe Ratio in the range direction before and after compensation for points A, B, and C. The changes in both the Peak-Lobe Ratio and Integrated Side-Lobe Ratio before and after compensation are minimal, with the Peak-Lobe Ratio hovering around -12dB, indicating good focusing of the target in the range direction itself.

[0182] Table 3 shows the azimuth peak sidelobe ratio and integral sidelobe ratio before and after compensation at each point.

[0183]

[0184] Table 4 shows the range-direction peak sidelobe ratio and integral sidelobe ratio before and after compensation at each point.

[0185]

[0186] Figure 11 The iterative curve of the sharpness value of the autofocus algorithm based on the maximum sharpness of the image is given. It can be seen from the iterative curve that the image sharpness increases with each iteration of the method. The image sharpness value increases from around 2.75E9 in the defocused image to around 5.74E9. It only needs three iterations to stabilize and obtain good focusing results. The method has fewer iterations, higher computational efficiency, and shorter time.

[0187] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this invention can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0188] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A method for bistatic ISAR backprojection imaging based on image autofocus, characterized in that, include: The radar echo signal is acquired, and the radar echo signal is processed using line frequency modulation demodulation to obtain a bistatic ISAR echo signal matrix; wherein, the radar echo signal includes: multiple pulse echoes; the size of the bistatic ISAR echo signal matrix is... , Indicates the number of range units of the echo. Indicates the number of azimuth units of the echo; The phase term in the bistatic ISAR echo signal matrix is ​​analyzed to obtain the translational and rotational components, and the translational components are compensated to obtain the translational compensation matrix. The dot product of the bistatic ISAR echo signal matrix and the translational compensation matrix is ​​performed to obtain the translationally compensated echo signal; wherein, the translationally compensated echo signal includes: translational error; The Doppler center is estimated and corrected on the translational compensation echo signal to eliminate the linear term in the translational error and obtain the first corrected echo signal. A BP imaging projection network is constructed based on the rotation component, and BP imaging is performed on the first corrected echo signal to obtain a BP imaging result; wherein, the BP imaging result is the sum of the back projection values ​​of multiple pulse echoes of the first corrected echo signal; The evaluation index and optimization model are determined, the optimization model is iteratively solved using the coordinate descent method, and the solution results are evaluated using the evaluation index. The phase estimation set is obtained based on the evaluation results. The first corrected echo signal is compensated according to the phase estimation set to obtain the corrected echo signal, and the corrected echo signal is subjected to BP imaging to obtain the final imaging result.

2. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The process of acquiring radar echo signals and processing them using line frequency modulation demodulation to obtain a bistatic ISAR echo signal matrix includes: The radar echo signal is compressed by range pulse compression using demodulation frequency modulation to obtain a compressed signal. The RVP phase term and envelope skew phase term of the compressed signal are compensated to obtain the bi-base ISAR echo signal matrix.

3. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The dual-base ISAR echo signal matrix is ​​represented as follows: ； in, This represents the scattering coefficient of the scattering point in the direction of the receiving radar, and its amplitude is... Phase is ; It is a slow time series. ; For fast time series, ; The center frequency of the transmitted signal; The pulse width of the transmitted signal; To adjust the frequency; This represents the sum of the distances from the two radars to the scattering point and the sum of the distances from the two radars to the reference point.

4. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The step involves analyzing the phase terms in the bistatic ISAR echo signal matrix to obtain translational and rotational components, and compensating for the translational components to obtain a translational compensation matrix, including: The initial phase term expression in the bistatic ISAR echo signal matrix is ​​determined, and the linearity of the radar echo signal is used to modify and derive the initial phase term expression to obtain the derived expression. The derived expression is decomposed to obtain the translational component and the rotational component; The translational components are compensated using the translational compensation method to obtain the translational compensation matrix.

5. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The translational error is expressed as: ； in, To slow time The relevant polynomial; To adjust the frequency; The center frequency of the transmitted signal; It is a slow time series. ; For fast time series, ; It is the speed of light.

6. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The step of constructing a BP imaging projection network based on the rotation component and performing BP imaging on the first corrected echo signal to obtain the BP imaging result includes: Determine the rotation component of the echo signal at any scattering point, and construct a BP imaging projection grid based on the rotation component; Perform a distance-dimensional Fourier transform on the first corrected echo signal to obtain the transform result; wherein the transform result is related to the rotation component; The transformation result is then subjected to BP imaging on the BP imaging projection grid to obtain the BP imaging result.

7. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 6, characterized in that, The transformation result is specifically expressed as follows: ； in, This represents the scattering coefficient of the scattering point in the direction of the receiving radar; Indicates the pulse width of the transmitted signal; Indicates frequency modulation; Represents the speed of light; Indicates the coherent difference frequency; It represents the sum of the rotational phase and the translational error phase; This indicates the center frequency of the transmitted signal.

8. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The BP imaging results are expressed as follows: ； in, This represents the result after the distance is transformed into a Fourier transform; Indicates the coherent difference frequency; It represents the rotation angle of the target relative to the equivalent monostatic radar over a slow time interval; Indicates the carrier frequency; Indicates the bibasic angle; It is a slow time series. ; Represents the projected grid coordinate; Represents the projected grid coordinate.

9. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The evaluation metrics include image sharpness, and the evaluation metrics are specifically expressed as follows: ； in, For the first Image intensity of each pixel; Represents the set of phase estimates; This indicates the number of distance units in the echo.

10. The method for bistatic ISAR backprojection imaging based on image autofocus according to claim 1, characterized in that, The step of compensating the first corrected echo signal according to the phase estimation set to obtain the corrected echo signal includes: A phase estimation set is obtained by using a self-focusing algorithm based on the maximum image sharpness; The phase compensation factor is obtained using the phase estimates from the aforementioned phase estimation set. The first corrected echo signal is compensated according to the phase compensation factor to obtain the corrected echo signal.

Images