Space target large-angle high-resolution radar imaging space-varying phase compensation method, system, storage medium and electronic device

Abstract

The application discloses a space target large-angle high-resolution radar imaging space-varying phase compensation method and system, a storage medium and an electronic device, and comprises the following steps: acquiring an ISAR original echo, completing a translation compensation and a distance beyond a cell correction; and converting residual two-dimensional space-varying phase error compensation into a high-dimensional parameter estimation problem based on image entropy minimization. A mean value, a step length and a covariance matrix of a CMA-ES algorithm are initialized, a candidate solution population is iteratively generated, and image entropy is calculated after compensation is performed on each group of space-varying phase coefficients to serve as fitness. An optimal individual is screened, a cumulative evolution path and a conjugate path are fused, and population parameters are adaptively updated. After a convergence condition is met, optimal coefficients are output, and a high-focusing ISAR image is reconstructed. The application overcomes the defects of a traditional gradient algorithm, such as being prone to local optimization and relying on initial values, utilizes global step length dynamic adjustment and a covariance adaptive mechanism of CMA-ES, has strong global optimization capability and noise robustness, and can still converge stably in a low signal-to-noise ratio environment.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing and space situational awareness (SSA) technology, and more specifically, to a method, system, storage medium, and electronic device for space target high-resolution radar imaging with large rotation angles and spatial phase compensation. Background Technology

[0002] In recent years, space situational awareness has played an increasingly important role in both military and civilian fields. Among them, inverse synthetic aperture radar (ISAR) imaging, due to its unique ability to provide high-resolution images of non-cooperative targets in all weather conditions and at all times, has become a core means of identifying and monitoring orbital assets in highly dynamic scenarios.

[0003] In air-to-air ISAR imaging scenarios, achieving extremely high azimuth resolution requires accumulating large radar observation angles. However, the accumulation of large angles inevitably exacerbates the complex, non-uniform three-dimensional rotational motion of the target relative to the radar. This non-uniform rotation leads to continuous changes in the effective rotation vector (ERV) and image projection plane (IPP), fundamentally causing severe two-dimensional spatially varied phase errors. These spatially varied distortions differ greatly between different scattering centers, causing traditional approximate compensation methods to fail, resulting in severe image defocusing and geometric distortion, and consequently significantly reducing the efficiency of subsequent target identification and classification.

[0004] Assuming the target's translational motion and cross-range cell (MTRC) have been pre-compensated, extracting unknown quadratic spatially varying coefficients is typically transformed into a high-dimensional parameter estimation problem based on minimum image entropy optimization. To solve this problem, existing techniques often employ gradient-based optimization algorithms, such as the Broyden-Fletcher-Goldfarb-Shanno (BFGS) or finite-memory BFGS (L-BFGS) algorithms for phase error coefficient search. However, these techniques suffer from significant drawbacks: due to unavoidable noise interference, the image entropy surface exhibits highly non-convex characteristics. Gradient-based solvers are extremely sensitive to initial values ​​and easily get trapped in local minima. To overcome this limitation, existing methods are forced to design extremely complex initial value estimation mechanisms (such as relying on extended Kalman filtering, radar tracking measurements, or ideal relative orbital geometry priors), which greatly limits the algorithm's engineering applicability. Even more seriously, in harsh electromagnetic environments such as low signal-to-noise ratio (SNR), the cost function surface becomes further distorted, and the gradient algorithm gets caught in a large number of invalid line searches, ultimately causing the algorithm to converge prematurely and phase compensation to fail completely. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, storage medium, and electronic device for space target high-resolution radar imaging with large rotation angles to compensate for spatial phase variation, thereby solving the technical defects of existing traditional gradient optimization algorithms that are prone to getting trapped in local optima under complex non-convex error surfaces and low signal-to-noise ratio environments, and that heavily rely on prior settings of initial values.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] This application provides a space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles. The method includes: Step S101, acquiring the raw ISAR echo data of the space target; Step S102, sequentially performing translational motion compensation and range cell correction processing on the acquired echo data to isolate the target's translational error; Step S103, transforming the residual two-dimensional space-varying phase error compensation (caused by the target's complex non-uniform rotation, resulting in nonlinear distortion in the range and azimuth directions depending on the spatial position, leading to non-uniform local defocusing and geometric deformation of the image) into a parameter estimation problem based on minimizing image entropy, and initializing a parameter set including the mean vector, global step size, and covariance matrix; Step S104, generating a population of several candidate solutions based on the current parameter set, and using the generated candidate solutions to perform independent space-varying phase error compensation operations on the ISAR data to reconstruct the image; Step S105... Step S106: Calculate the image entropy value for each set of image data generated after the above compensation, and use it as a fitness index to evaluate the image focusing quality and parameter quality; Step S107: Select the best individuals in the population based on the calculated image entropy value, and adaptively update the mean vector, global step size and covariance matrix of the next generation population by combining the cumulative evolution path and the conjugate path; Step S108: Enter the iteration process (the parameters initialized in step S103 are used in the first execution, and the parameters updated in step S106 are used in subsequent iterations), and determine whether the image entropy difference between adjacent iterations and the current global step size meet the preset minimum convergence condition. If not, return to step S109; Step S109: If the convergence condition is met, extract the optimal spatially variable phase coefficient estimate and use the parameter to reconstruct and generate the final high-focus ISAR image.

[0008] Preferably, in step S101, acquiring the raw ISAR echo data of the space target specifically involves: the radar transmitting a linear frequency modulated signal, and the raw echo signal intercepted by the receiver at any scattering point P on the target is represented in the time domain as:

[0009]

[0010] in, Represents a rectangular window function. It is the frequency modulation slope of the linear frequency modulation signal. Represents the speed of light. It is the carrier frequency. Represents pulse width. Represents distance in time. Represents slow time The instantaneous radial distance between the radar and the scattering point P.

[0011] Preferably, in step S102, after translational motion compensation and MTRC correction processing, the residual two-dimensional spatially variable phase error signal can be expressed as follows: After sequentially performing translational motion compensation and range-crossing cell correction processing on the echo data, the residual two-dimensional spatially variable phase error signal in the range-frequency domain-azimuth time domain can be expressed as follows:

[0012]

[0013] in, This represents an ideal discrete signal with no phase error. Represents the wavelength of the radar transmitted signal. Represents the range frequency. and These represent the space-variable parameter mapping vectors for range and azimuth, respectively. and The constant term and coefficient of the first-order term for the distance change caused by the rotational motion of the target.

[0014] Preferably, in step S103, the purpose of initializing the parameter set is to transform the residual two-dimensional spatially variable phase error compensation into a parameter estimation problem based on minimizing image entropy:

[0015]

[0016] in, Represents the image entropy evaluation function. and These are the azimuth space variation coefficients and range space variation coefficients to be compensated, respectively, and their physical definitions are as follows:

[0017]

[0018]

[0019] in, and These are spatially variable parameter mapping vectors for the azimuth and range directions, respectively. The initialization settings also include: an initial mean vector. Initial global step size and the initial covariance matrix To accommodate the differences in sensitivity to phase errors in different directions.

[0020] Preferably, in step S104, generating candidate solutions specifically involves:

[0021]

[0022] in, Represents an individual index. Represents population size, Represents the current iteration number. The generated candidate solution vector, This is the current average. The current step size, To conform to a multivariate normal distribution A random vector. Using the spatially varying phase error step size of each candidate solution, the image is reconstructed. It can be represented as:

[0023]

[0024] in, Represents aperture index, Represents the total number of apertures in the azimuth direction. Represents the directional index. This represents a preprocessed discrete signal. represents the spatially variable coefficients in the candidate solutions.

[0025] Preferably, in step S105, the calculation of the image entropy value specifically involves:

[0026]

[0027] in, Represents the total energy of the image. This represents the total number of distance units.

[0028] Preferably, in step S106, updating the distribution parameter model specifically includes using the previous Update the mean vector for each preferred individual:

[0029]

[0030] in To reorganize the weights, construct a cumulative evolution path. To dynamically update the global step size To prevent the algorithm from getting stuck in a local entropy minimum and to adjust the search speed, the specific formula is as follows:

[0031]

[0032]

[0033] in and These represent the learning rate and damping parameter, respectively. The expected value representing the length of a specific random variable. Contribute weights to the effective variance. This is the weighted average step size for the optimal individual. It is calculated as follows:

[0034]

[0035] Synergistic conjugate path The covariance matrix is ​​adaptively updated with the optimal population variance, where the conjugate matrix... The update formula is:

[0036]

[0037] in, The learning rate represents the conjugate path. The step function used to control path updates, combined with these update strategies, leads to the final update formula for the covariance matrix:

[0038]

[0039] in, and Level 1 and Level 2 respectively Update the specific learning rate.

[0040] Preferably, in step S107, the convergence condition specifically involves determining whether the image entropy difference between adjacent iterations and the current global step size simultaneously satisfy a preset threshold.

[0041]

[0042]

[0043] Preferably, in step S108, the output of the final focused ISAR image specifically involves: if the conditions described in step S107 are met, the iteration is terminated, and the optimal parameter estimate at this time is extracted. The optimal spatially variable phase coefficient is used to finally compensate for the residual two-dimensional spatially variable phase error, and a high-fidelity, high-focus two-dimensional ISAR image of the target is reconstructed and output.

[0044] This application also provides a space-varying phase compensation system for high-resolution radar imaging of space targets with large rotation angles, including:

[0045] The signal acquisition unit is configured to acquire the raw ISAR echo data of the space target; the preprocessing unit is configured to sequentially perform translational motion compensation and cross-range unit correction processing on the acquired echo data to isolate the translational error of the target and obtain the signal to be compensated.

[0046] The parameter initialization unit is configured to transform the residual two-dimensional spatially variable phase error compensation into a parameter estimation problem based on minimizing image entropy, and initialize a parameter set including the mean vector, global step size, and covariance matrix.

[0047] The compensation and reconstruction unit is configured to generate a population consisting of several candidate solutions based on the current parameter set, and to use the generated candidate solutions to perform independent space-varying phase error compensation operations on the signal to be compensated to reconstruct and generate image data.

[0048] The quality assessment unit is configured to calculate the image entropy value for each set of image data generated after compensation, and use it as a fitness index to evaluate the image focusing quality and parameter quality; the evolution update unit is configured to select the best individuals in the population based on the calculated image entropy value, and adaptively update the mean vector, global step size and covariance matrix of the next generation population by combining the cumulative evolution path and the conjugate path.

[0049] The convergence determination and output unit is configured to determine whether the image entropy difference between adjacent iterations and the current global step size simultaneously satisfy the preset minimum convergence condition; if not, the compensation and reconstruction unit is triggered to continue iterating; if satisfied, the optimal spatially variable phase coefficient estimate is extracted, and the parameter is used to reconstruct the final high-focus ISAR image.

[0050] This application also provides an electronic device, including a memory and a processor; the memory is used to store a computer program; the processor is coupled to the memory and is used to execute the computer program stored in the memory to implement the CMA-ES-based air-to-air large-angle ISAR imaging spatial phase compensation method as described above.

[0051] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the CMA-ES-based air-to-air large-angle ISAR imaging spatial phase compensation method as described above.

[0052] Beneficial effects:

[0053] This application provides a space-varying phase compensation method for high-resolution radar imaging of space targets at large rotation angles. Firstly, by abstracting the ISAR image phase correction process into a high-dimensional non-convex optimization problem, the global optimization characteristic of the CMA-ES algorithm is utilized to effectively solve the technical problem that traditional gradient-based algorithms (such as BFGS) are prone to getting trapped in local optima when dealing with highly non-convex error surfaces caused by non-uniform rotation. Secondly, this invention introduces an adaptive covariance matrix update and cumulative step size control mechanism, eliminating the need for high-precision initial value prior settings and greatly reducing the system's dependence on radar tracking accuracy and orbital geometry prior information. Furthermore, while maintaining the same order of magnitude of computational complexity as traditional algorithms, this invention significantly improves robustness in low signal-to-noise ratio (SNR) environments, maintaining stable convergence performance even under extremely low SNR conditions. It can effectively eliminate azimuth defocus caused by large rotation angle observations and accurately reconstruct the subtle electromagnetic scattering characteristics of the target. This method has significant engineering application value for improving the existing high-resolution ISAR imaging technology system for space targets and enhancing the effectiveness of space situational awareness. Attached Figure Description

[0054] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0055] Figure 1 This is an overall flowchart of the space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles provided in an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of the space geometric observation model for air-to-air ISAR imaging of space targets according to an embodiment of the present invention;

[0057] Figure 3 This is a schematic diagram of the satellite point target distribution model involved in an embodiment of the present invention;

[0058] Figure 4 This is a comparison chart of the point target imaging performance of the embodiments of the present invention and existing traditional algorithms under different signal-to-noise ratio environments during the first observation period;

[0059] Figure 5 This is a comparison chart of the point target imaging performance of the embodiment of the present invention and the existing traditional algorithm under different signal-to-noise ratio environments during the second observation period;

[0060] Figure 6 This is a graph showing the three-dimensional distribution of the image entropy surface in an embodiment of the present invention and a comparison of the convergence performance of different algorithms. Detailed Implementation

[0061] The present application will now be described in detail with reference to the accompanying drawings and embodiments. Various examples are provided by way of explanation and not by way of limitation. In fact, those skilled in the art will recognize that modifications and variations can be made to the present application without departing from the scope or spirit thereof. For example, a feature shown or described as part of one embodiment may be used in another embodiment to produce yet another embodiment. Therefore, it is desirable that the present application encompass such modifications and variations that fall within the scope of the appended claims and their equivalents.

[0062] Exemplary methods

[0063] like Figure 1 As shown, the space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles includes:

[0064] In step S101, the raw ISAR echo data of the space target is acquired, specifically: the radar transmits a linear frequency modulated signal, and the raw echo signal intercepted by the receiver at any scattering point P on the target is represented in the time domain as:

[0065]

[0066] in, Represents a rectangular window function. It is the frequency modulation slope of the linear frequency modulation signal. Represents the speed of light. It is the carrier frequency. Represents pulse width. Represents distance in time. Represents slow time The instantaneous radial distance between the radar and the scattering point P.

[0067] In step S102, after translational motion compensation and MTRC correction, the residual two-dimensional spatially variable phase error signal can be expressed as follows: After sequentially performing translational motion compensation and range-crossing cell correction on the echo data, the residual two-dimensional spatially variable phase error signal in the range-frequency domain-azimuth time domain can be expressed as follows:

[0068]

[0069] in, This represents an ideal discrete signal with no phase error. Represents the wavelength of the radar transmitted signal. Represents the range frequency. and These represent the space-variable parameter mapping vectors for range and azimuth, respectively. and The constant term and coefficient of the first-order term for the distance change caused by the rotational motion of the target.

[0070] In step S103, the purpose of initializing the parameter set is to transform the residual two-dimensional spatially variable phase error compensation into a parameter estimation problem based on minimizing image entropy:

[0071]

[0072] in, Represents the image entropy evaluation function. and These are the azimuth space variation coefficients and range space variation coefficients to be compensated, respectively, and their physical definitions are as follows:

[0073]

[0074]

[0075] in, and These are spatially variable parameter mapping vectors for the azimuth and range directions, respectively. The initialization settings also include: an initial mean vector. Initial global step size and the initial covariance matrix To accommodate the differences in sensitivity to phase errors in different directions.

[0076] In step S104, generating candidate solutions specifically involves:

[0077]

[0078] in, Represents an individual index. Represents population size, Represents the current iteration number. The generated candidate solution vector, This is the current average. The current step size, To conform to a multivariate normal distribution A random vector. Using the spatially varying phase error step size of each candidate solution, the image is reconstructed. It can be represented as:

[0079]

[0080] in, Represents aperture index, Represents the total number of apertures in the azimuth direction. Represents the directional index. This represents a preprocessed discrete signal. represents the spatially variable coefficients in the candidate solutions.

[0081] In step S105, the calculation of the image entropy value is specifically as follows:

[0082]

[0083] in, Represents the total energy of the image. This represents the total number of distance units.

[0084] In step S106, updating the distribution parameter model specifically includes using the previous... Update the mean vector for each preferred individual:

[0085]

[0086] in To reorganize the weights, construct a cumulative evolution path. To dynamically update the global step size To prevent the algorithm from getting stuck in a local entropy minimum and to adjust the search speed, the specific formula is as follows:

[0087]

[0088]

[0089] in and These represent the learning rate and damping parameter, respectively. The expected value representing the length of a specific random variable. Contribute weights to the effective variance. This is the weighted average step size for the optimal individual. It is calculated as follows:

[0090]

[0091] Synergistic conjugate path The covariance matrix is ​​adaptively updated with the optimal population variance, where the conjugate matrix... The update formula is:

[0092]

[0093] in, The learning rate represents the conjugate path. The step function used to control path updates, combined with these update strategies, leads to the final update formula for the covariance matrix:

[0094]

[0095] in, and Level 1 and Level 2 respectively Update the specific learning rate.

[0096] In step S107, the convergence condition is specifically: determining whether the image entropy difference between adjacent iterations and the current global step size simultaneously satisfy a preset threshold.

[0097]

[0098]

[0099] In step S108, the final focused ISAR image is output as follows: if the conditions described in step S107 are met, the iteration is terminated, and the optimal parameter estimate at this time is extracted. The optimal spatially variable phase coefficient is used to finally compensate for the residual two-dimensional spatially variable phase error, and a high-fidelity, high-focus two-dimensional ISAR image of the target is reconstructed and output.

[0100] The spatial geometric observation model for air-to-air ISAR imaging of space targets involved in this embodiment is as follows: Figure 2 As shown. The radar payload is deployed in Low Earth Orbit (LEO), and the target is a non-cooperative satellite in an adjacent orbit. The main system parameters used in the experiment are: carrier frequency of 35 GHz, signal bandwidth of 3 GHz, observation rotation angle of 8°, and coherent processing pulse number of 256. The point target distribution model of the target satellite is shown below. Figure 3 As shown. Figure 4 and Figure 5 As shown, this embodiment compares the imaging results of the method of the present invention with those of the traditional RD algorithm, PFA algorithm, and gradient-based traditional optimization algorithms (such as BFGS) at two different observation periods. Figure 4 (a) and Figure 5 As shown in (a), the RD algorithm can only compensate for spatial invariance errors and cannot correct for cross-range cells, causing the target features to degenerate into blurry oblique bands at all signal-to-noise ratios. Figure 4 (b) and Figure 5 As shown in (b), although the PFA algorithm can initially reconstruct the target outline, it neglects the two-dimensional spatial variation phase error, resulting in severe azimuth diffusion at the scattering center of the solar panel, appearing as an elongated dashed line. Figure 4 (c) and Figure 5 As shown in (c), although traditional gradient optimization algorithms have shown some improvement, they are still limited by the local optima trap of non-convex entropy surfaces, and residual fuzziness remains. Figure 4 (d) and Figure 5 As shown in (d), the CMA-ES-based method provided by this invention can accurately estimate two-dimensional spatially variable parameters, significantly suppress azimuth defocus, and reconstruct solar panel details into clear, independent point structures. This embodiment analyzes the surface features of the image entropy evaluation function. For example... Figure 6 As shown in (a), the image entropy surface exhibits highly non-convex characteristics in the presence of noise interference. Figure 6 (b) shows that the convergence curve indicates that traditional gradient algorithms are prone to getting trapped in local minima, leading to search failure. In contrast, the method of this invention, through adaptive adjustment of the covariance matrix, can successfully bypass local barriers and achieve robust global convergence. Even in complex electromagnetic scattering environments, the method of this invention can still clearly reconstruct the satellite's body and complex solar panel structure, exhibiting stronger robustness and imaging fidelity compared to traditional methods. This provides strong support for ultra-high resolution ISAR imaging of high-dynamic non-cooperative targets and space situational awareness.

[0101] Exemplary System

[0102] This application embodiment also provides a space-varying phase compensation system for high-resolution radar imaging of space targets with large rotation angles, including: a signal acquisition unit configured to acquire raw ISAR echo data of a space target; specifically, receiving the target's scattered echo using a radar antenna and performing down-conversion and orthogonal demodulation processing to obtain a raw discrete echo sequence. A motion compensation and correction unit configured to sequentially perform translational motion compensation and range-crossing unit correction processing on the echo data to isolate the target's translational motion error and eliminate range migration, obtaining the residual phase error signal to be processed. A parameter initialization unit configured to initialize the distribution parameter set of the covariance matrix adaptive evolution strategy (CMA-ES); specifically, setting an initial mean vector. Initial global step size And the anisotropic initial covariance matrix configured for the differences between the azimuth and range directions. The candidate solution generation and compensation unit is configured to generate a population of multiple candidate solutions based on the current distribution parameter set, and use each candidate solution to perform two-dimensional spatial phase compensation on the residual phase error signal to reconstruct multiple sets of images to be evaluated. The image entropy evaluation unit is configured to calculate the image entropy value of each set of images to be evaluated, and use the image entropy value as a fitness index to evaluate the imaging focusing quality and the merits of the parameter solutions. The distribution model evolution and update unit is configured to select the best individuals based on the image entropy value, and adaptively update the mean vector, global step size, and covariance matrix of the next generation population by combining the cumulative evolution path and the conjugate path, so that the search distribution evolves in the direction of decreasing entropy value. The convergence determination and iteration unit is configured to determine whether the image entropy difference between adjacent iterations and the current global step size have both reached a preset convergence threshold; if not, the candidate solution generation and compensation unit is triggered to enter the next round of evolution iteration. Final image output unit: configured to output the optimal spatially varying phase coefficient estimate when the convergence condition is met, and reconstruct the final high-focus, high-fidelity 2D ISAR image accordingly.

[0103] Exemplary device

[0104] A computer-readable storage medium stores a computer program thereon. When executed by a processor, the computer program causes the device containing the computer-readable storage medium to perform the CMA-ES-based air-to-air large-angle ISAR imaging spatially variable phase compensation method as described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), random access memory, and other memories.

[0105] An electronic device includes a memory and a processor, wherein the memory stores a program executable on the processor, and the processor executes the program to implement the CMA-ES-based air-to-air large-angle ISAR imaging spatial phase compensation method as described above.

[0106] If the modules / units integrated in the electronic device described in this application are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can also be implemented by a computer program instructing related hardware devices. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above.

[0107] Furthermore, the computer-readable storage medium may primarily include a stored program area and a stored data area, wherein the stored program area may store the operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of ISAR imaging processing, etc.

[0108] The computer-readable storage medium stores computer-readable instructions, which are executed by a processor in an electronic device to implement the space-variable phase compensation method for high-resolution radar imaging of space targets with large turning angles as described in any of the above embodiments.

[0109] This application can robustly overcome local optima traps under harsh signal-to-noise ratio environments and complete two-dimensional spatial phase compensation caused by large-angle non-uniform rotation, significantly improving the reconstruction quality of high-resolution images of space targets. It has important engineering application significance for improving the existing space situational awareness (SSA) technology system.

[0110] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0111] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0112] The antenna, including monopulse antennas and phased array antennas, is one of the most critical components in the radar system. The memory includes high-speed, high-capacity solid-state memory or digital radio frequency memory, etc. The processor includes a central processing unit (CPU), a network processor (NP), etc., and can also be a digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), control circuit, or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0113] The antenna can be specifically configured to transmit linear frequency modulated radar signals to the space target to be observed and receive the raw ISAR echo signals reflected by the target.

[0114] Specifically, the processor can be configured to: sequentially perform translational motion compensation and cross-range cell correction processing on the captured ISAR raw echo signal to isolate the target's translational error; transform the residual two-dimensional spatially varied phase error compensation into a parameter estimation problem based on minimizing image entropy, and initialize the distribution parameter set of the covariance matrix adaptive evolution strategy (CMA-ES); generate a population composed of multiple candidate solutions through iteration, and perform independent spatially varied phase error compensation operations on the echo data respectively; calculate the image entropy value of the reconstructed image to select the best individuals; adaptively update the mean vector, global step size, and covariance matrix of the next generation population by combining the evolution path and the conjugate path; when the preset convergence condition is met, terminate the iteration and extract the optimal spatially varied phase coefficient estimate, and reconstruct a high-fidelity two-dimensional focused ISAR image of the target.

[0115] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0116] The methods described above according to the embodiments of this application can be implemented in hardware, software, or a combination of hardware and software. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory (such as a CD-ROM, RAM, hard disk, or magnetic disk), or implemented as a raw storage medium downloaded over a network, located on a remote recording medium or a non-transitory machine storage medium, and have appropriate instructions directed to the system and methods for execution. Furthermore, when a general-purpose computer accesses the code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the methods shown herein.

[0117] Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions or included in processor control code. Whether these functions are implemented in hardware, software, or a combination of both depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the embodiments of this application.

[0118] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0119] The device and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0120] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A space-varying phase compensation method for high-resolution radar imaging of space targets at large rotation angles, characterized in that, Includes the following steps: Step S101: Obtain raw ISAR echo data of space targets; Step S102: Perform translational motion compensation and cross-range cell correction processing sequentially on the raw ISAR echo data; Step S103: Initialize the global distribution parameters of the covariance matrix adaptive evolution strategy algorithm. The global distribution parameters include the initial mean vector, the global step size, and the initial covariance matrix. The residual two-dimensional spatially variable phase error compensation is transformed into a parameter estimation problem based on the minimization of image entropy. Step S104: First, based on the current mean vector, global step size and covariance matrix, a population of several candidate solutions is generated by random sampling through a multivariate normal distribution, where each candidate solution corresponds to a set of two-dimensional spatially variable phase error coefficients to be searched; Second, each candidate solution in the population is extracted as an independent compensation parameter, and two-dimensional spatially variable phase error compensation is performed on the ISAR data processed in step S102 respectively, and multiple ISAR images to be evaluated corresponding to the population are reconstructed. Step S105: Calculate the image entropy for each set of image data generated after compensation in step S104, and use the image entropy as a fitness index to evaluate the image focusing quality and parameter quality. Step S106: Based on the calculation results of step S105, select the optimal number of individuals with the minimum image entropy from the population; fuse the cumulative evolution path and the conjugate path to adaptively update the mean vector, global step size and covariance matrix of the next generation population. Step S107: Repeat steps S104-S106 to enter the parameter iterative optimization process and determine whether the image entropy difference between adjacent iterations and the current global step size simultaneously meet the preset minimum convergence condition. Step S108: Determine whether the condition in S107 is met. If not, increase the number of iterations and return to step S104. If the condition is met, output the optimal spatially variable phase coefficient estimate and use the coefficient to reconstruct the final high-focus ISAR image.

2. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, In step S101, acquiring the raw ISAR echo data of the space target specifically involves: the radar transmitting a linear frequency modulated signal, and the raw echo signal intercepted by the receiver at any scattering point P on the target is represented in the time domain as: ； in, Represents the imaginary unit. It is the scattering coefficient. Represents a rectangular window function. It is the frequency modulation slope of the linear frequency modulation signal. Represents the speed of light. It is the carrier frequency. Represents pulse width. Represents distance in time. Represents slow time The instantaneous radial distance between the radar and the scattering point P.

3. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, In step S102, after sequentially performing translational motion compensation and range-crossing cell correction on the echo data, the signal with residual two-dimensional spatially variable phase error can be expressed in the range-frequency domain-azimuth time domain as follows: ； in, This represents an ideal discrete signal with no phase error. Represents the wavelength of the radar transmitted signal. Represents the range frequency. and These represent the space-variable parameter mapping vectors for the range and azimuth directions, respectively. and The constant term and coefficient of the first-order term for the distance change caused by the rotational motion of the target.

4. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, In step S104, the generation of candidate solutions in the population is specifically represented as follows: ； in, Represents an individual index. Represents population size, Represents the current iteration number. The generated candidate solution vector, This is the current average. The current step size, The image is reconstructed using a random vector that follows a multivariate normal distribution and a spatially varying phase error step size derived from each candidate solution. It can be represented as: ； in, Represents aperture index, Represents the total number of apertures in the azimuth direction. Represents direction to all This represents a preprocessed discrete signal. represents the spatially variable coefficients in the candidate solutions.

5. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, In step S105, the image entropy value corresponding to each candidate solution is calculated as follows: ； in, Represents the total energy of the image. The total number of distance cells. This represents the total number of azimuth units. This represents the discrete pixel values ​​of the reconstructed image.

6. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, The updated distribution parameter model mentioned in step S106 specifically includes the selected top Update the mean vector for each preferred individual: ； in To reorganize the weights, Represents the [number]th [rank] in the current population fitness ranking after sorting by fitness. The solution vector of the spatially variable phase coefficients corresponding to the optimal candidate individuals. This represents the current iteration number.

7. The space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles according to claim 1, characterized in that, In step S107, the convergence condition is specifically as follows: ； ； in, and Representing the first Second and third The optimal image entropy value of the reconstructed image after the nth iteration. This represents the preset convergence threshold for image entropy difference; Representing the The current global step size in the next iteration. This represents the preset global step size minimum convergence threshold; if the convergence condition is met, proceed to step S108; otherwise, proceed to step S104.

8. A space-varying phase compensation system for high-resolution radar imaging of space targets with large rotation angles, characterized in that, include: The data acquisition and preprocessing unit is configured to acquire the raw ISAR echo data of the space target and sequentially perform translational motion compensation and cross-range unit correction processing on the echo data. The parameter initialization unit is configured to transform the residual two-dimensional spatially variable phase error compensation into a parameter estimation problem based on image entropy minimization, and initialize the parameter set, which includes an initial mean vector, a global step size, and an initial covariance matrix. The solution space generation and compensation unit is configured to generate a population consisting of several candidate solutions based on the current parameter set, and to perform independent two-dimensional spatial variation phase error compensation operations on the preprocessed ISAR data using each of the generated candidate solutions. The image entropy evaluation unit is configured to calculate the image entropy for each set of image data generated after compensation, and use the image entropy as a fitness index to evaluate the image focusing quality and parameter quality. The distribution model update unit is configured to select preferred individuals from the population based on the image entropy value, and integrate the cumulative evolution path and the conjugate path to adaptively update the mean vector, global step size and covariance matrix of the next generation population. The convergence determination and output unit is configured to determine whether the image entropy difference between adjacent iterations and the current global step size meet the preset minimum convergence condition. If not, the solution space generation and compensation unit is triggered to perform the next iteration. If it meets the condition, the optimal spatially variable phase coefficient estimate is output, and the final focused ISAR image is reconstructed using the coefficient.

9. An electronic device comprising a memory and a processor; the memory for storing a computer program; the processor, coupled to the memory, for executing the computer program stored in the memory to implement the space-varying phase compensation method for high-resolution radar imaging of space targets with large rotation angles as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the space-variable phase compensation method for high-resolution radar imaging of space targets with large rotation angles as described in any one of claims 1-7.

Images