Space target on-orbit microwave imaging method for decoupling of translation and rotation based on parameter estimation

Abstract

The application relates to the technical field of space imaging, and provides a spatial target space-based microwave imaging translation-rotation decoupling method based on parameter estimation. The method first compensates a translation motion component of microwave imaging echo data by using an adjacent cross-correlation method and a self-focusing method, then optimizes and compensates a translation linear phase component by combining a Keystone transformation with a first-order parameter of the translation motion fitted by the former and taking the minimum image entropy of echo data as a cost function; subsequently, the method compensates a translation motion component caused by a space-varying distance cell translation of a rotation motion component, then performs residual translation motion component compensation again, obtains echo data after the translation component and the rotation MTRC are basically completely compensated, and performs space-varying phase error compensation and azimuth compression on the echo data, so that the imaging result of the spatial target is obtained, and the efficiency and accuracy of parameter estimation are considered, and the imaging quality is effectively improved.

Description

Technical Field

[0001] This application relates to the field of space imaging technology, and in particular to a method for decoupling translational rotation of space target space-based microwave imaging based on parameter estimation. Background Technology

[0002] Inverse-synthetic-aperture radar (ISAR, microwave imaging) is an important implementation method of microwave imaging. As a remote sensing imaging method for moving targets, especially non-cooperative moving targets, ISAR has the characteristics of all-day, all-weather operation and no limitation on the operating range. It is currently the main observation method for air and space targets. Its core lies in achieving high lateral resolution by utilizing the Doppler frequency difference generated by the mutual motion between the scattering points at different positions of the target and the radar.

[0003] In microwave imaging, the motion of a target is generally decomposed into translation and rotation for processing and compensation. Translational compensation eliminates the influence of uniform target motion. In this case, the relative rotation between the two can be equivalent to the target's rotation around its center of mass. The Doppler between each scattering point and the radar can be considered a constant that depends only on the azimuth position of the scattering point, and azimuth focusing can be achieved with a first azimuth Fourier transform. However, with the continuous improvement of radar system resolution and the increasing demand for large-angle observation in imaging scenarios, under high-resolution, large-angle conditions, the target's rotation not only produces Doppler frequency shift but also introduces non-negligible Migration Through Resolution Rell (MTRC) and spatially varied phase errors.

[0004] However, existing microwave imaging methods treat translation and rotation separately, neglecting the close coupling between the two in echo data at high resolution and large rotation angles. Rotational errors affect the accuracy of traditional translational compensation methods, and in cascaded processing frameworks, errors in the translational compensation stage inevitably propagate and amplify to the rotational compensation stage, severely undermining the fundamental assumptions upon which traditional methods rely. Although some techniques now consider joint translational and rotational compensation, incorporating translational and rotational parameters into a high-dimensional optimization problem for solution, these methods often encounter new problems when performing high-resolution space-based microwave imaging of space targets, such as model misfit, high computational complexity, and susceptibility to local optima. Summary of the Invention

[0005] In view of this, this application provides a space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation to solve the problems of low efficiency and insufficient accuracy of microwave imaging translational and rotational compensation in the prior art.

[0006] A first aspect of this application provides a method for decoupling translational rotation of space target space-based microwave imaging based on parameter estimation, comprising:

[0007] Acquire microwave imaging echo data after pulse compression; the echo data includes at least the range history, and the range history includes at least translational motion components and rotational motion components.

[0008] Preliminary compensation is performed on the translational motion components. This preliminary compensation includes: using the adjacent cross-correlation method to align the envelope of the echo data and determining the first-order translational motion parameters of the translational motion components based on the envelope alignment results; using the self-focusing method to estimate the global phase error and using the obtained global phase error to compensate for the azimuth data of the echo data.

[0009] The translational motion components are optimized and compensated. The optimization and compensation include: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining the Keystone transform to optimize the first-order translational motion parameters, thereby obtaining the optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational linear phase components.

[0010] The Keystone transform is used to compensate for the spatial displacement unit (MTRC) caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component.

[0011] The echo data after rotational MTRC compensation is then compensated again for translational residual motion components to obtain echo data after complete compensation of translational components and rotational MTRC.

[0012] Spatial phase error compensation and azimuth compression are performed on the echo data after full compensation of translational and rotational MTRC components to obtain the imaging results of the space target.

[0013] A second aspect of this application provides a space target space-based microwave imaging translational-rotation decoupling device based on parameter estimation, comprising:

[0014] The acquisition module is configured to acquire microwave imaging echo data after pulse compression; the echo data includes at least the distance history, and the distance history includes at least translational motion components and rotational motion components.

[0015] The translational compensation module is configured to perform preliminary compensation for the translational motion components. The preliminary compensation for the translational motion components includes: using the adjacent cross-correlation method to perform envelope alignment on the echo data, and determining the first-order translational motion parameters of the translational motion components based on the envelope alignment results; using the self-focusing method to estimate the global phase error, and using the obtained global phase error to compensate for the azimuth data of the echo data.

[0016] The translational compensation module is also configured to optimize and compensate the translational motion components; wherein, the optimization and compensation includes: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining the Keystone transform to optimize the first-order translational motion parameters, thereby obtaining the optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational linear phase components.

[0017] The rotation compensation module is configured to use Keystone transform to compensate for the space-time distance cell migration (MTRC) caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component.

[0018] The translational compensation module is also configured to perform translational residual motion component compensation again on the echo data after rotational MTRC compensation, so as to obtain echo data after complete compensation of translational components and rotational MTRC.

[0019] The imaging module is configured to perform spatially variable phase error compensation and azimuth compression on the echo data after full compensation of translational and rotational MTRC components, so as to obtain the imaging results of the space target.

[0020] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0021] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0022] The beneficial effects of the embodiments in this application compared with the prior art are:

[0023] This embodiment first uses the adjacent cross-correlation method and the self-focusing method to perform preliminary compensation for the translational motion component of the microwave imaging echo data after pulse compression. Then, using the first-order translational motion parameter fitted by the former as the initial value and the minimum image entropy of the echo data as the cost function, the Keystone transform is combined to optimize the compensation of the translational linear phase component. Next, the Keystone transform is used to compensate for the spatially varied distance cell migration caused by the rotational motion component. Then, the translational residual motion component is compensated again to obtain echo data after the translational component and rotational MTRC are basically fully compensated. The echo data after envelope alignment is then subjected to spatially varied phase error compensation and azimuth compression to obtain the imaging result of the space target. This achieves both efficiency and accuracy in parameter estimation, avoids the geometric distortion caused by the insensitivity of the first-order coefficients in traditional translational phase compensation methods in subsequent compensation, and effectively improves the imaging quality. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating a space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation, provided in an embodiment of this application.

[0026] Figure 2 This is a schematic diagram of the MTRC compensation results obtained using traditional methods.

[0027] Figure 3 This is a schematic diagram of the MTRC compensation result obtained using the technical solution provided in the embodiments of this application.

[0028] Figure 4 This is a schematic diagram of the imaging results obtained using traditional methods.

[0029] Figure 5 This is a schematic diagram of the imaging results obtained using the technical solution provided in the embodiments of this application.

[0030] Figure 6 This is a schematic diagram of the structure of a space-based microwave imaging translational-rotation decoupling device for space targets based on parameter estimation, provided in an embodiment of this application.

[0031] Figure 7 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0032] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0033] The following describes in detail, with reference to the accompanying drawings, a method and apparatus for decoupling translational rotation of space target space-based microwave imaging based on parameter estimation, according to embodiments of this application.

[0034] As mentioned above, with the continuous improvement of radar system resolution and the increasing demand for large-angle observation in imaging scenarios, under high-resolution and large-angle conditions, the rotation of the target not only produces Doppler frequency shift, but also introduces non-negligible MTRC and spatially variable phase errors.

[0035] To improve imaging quality, in 2021, Jin Wei et al. proposed Radon-Non-Uniform Fractional Fourier Transform (Radon-NUFrFT) to estimate motion parameters and used a search algorithm based on minimizing entropy to optimize the parameters, compensating for nonlinear MTRC and higher-order phase. In 2023, Xiaoyu Yang et al. proposed a novel rotational motion compensation algorithm, using the minimum residual norm of the basis signal phase and employing an iterative method to independently estimate different rotation parameters, avoiding problems caused by accumulated errors. In 2024, Can Liu et al. proposed a parameterized non-search method with two main stages, which has higher accuracy and robustness for translational compensation of non-stationary motion of targets under low signal-to-noise ratio conditions. In the same year, Jishun... Li et al. proposed a joint microwave imaging motion compensation algorithm based on entropy minimization under low SNR conditions. The motion of the target is modeled as a high-order polynomial, and a parameterized joint compensation model for high-speed motion and translational motion is established. In 2025, Wang Yong et al. proposed a microwave imaging envelope alignment method based on complex domain convolutional neural network, which improves the accuracy and computational efficiency of translational compensation through deep learning strategy. In the same year, Hou Qingsen et al. proposed a space target microwave imaging method based on fast estimation of joint motion parameters. The remaining translational phase error and rotational parameters are jointly estimated based on minimum entropy, minimizing the influence of the translational residual phase on the imaging results.

[0036] Existing technical methods mainly improve the efficiency and accuracy of translational and rotational compensation by combining different motion models, extract more detailed motion information, achieve more accurate data compensation, and improve imaging quality.

[0037] However, existing microwave imaging methods treat translation and rotation separately, neglecting the close coupling between the two in echo data at high resolution and large rotation angles. Rotational errors affect the accuracy of traditional translational compensation methods, and in cascaded processing frameworks, errors in the translational compensation stage inevitably propagate and amplify to the rotational compensation stage, severely undermining the fundamental assumptions upon which traditional methods rely. Although some techniques now consider joint translational and rotational compensation, incorporating translational and rotational parameters into a high-dimensional optimization problem for solution, these methods often encounter new problems when performing high-resolution space-based microwave imaging of space targets, such as model misfit, high computational complexity, and susceptibility to local optima.

[0038] In view of this, the embodiments of this application provide a space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation. Using minimum entropy as the benchmark, the translational motion parameters are searched in conjunction with rotational MTRC compensation. By combining two translational compensations, decoupling / high-resolution imaging of translation and rotation is achieved.

[0039] Specifically, this embodiment first employs traditional cross-correlation and phase gradient autofocus (PGA) for translational compensation. In this case, the translational residual component is mixed with the rotational MTRC (Mean Transformation-Modulation Reflectance). When using the Keystone Transform (KT), both are simultaneously distorted. In the translational residual, since autofocus is insensitive to linear phase estimation errors, this term is the main source of error. The first-order translational motion parameters are extracted from the envelope alignment data as initial values. Using minimum entropy as a benchmark, the optimal first-order translational motion parameters are searched using first-order translational phase error compensation and Keystone Transform to minimize the first-order translational phase error. After phase and spatially variable MTRC compensation using the obtained results, the remaining translational error still causes MTRC and azimuth defocus. Another translational compensation can be performed to minimize the translational residual error. Since the spatially variable MTRC caused by large rotation angles can be considered basically corrected at this point, the accuracy of translational compensation is further improved when there is no rotational MTRC component. At this point, translational compensation can be considered basically complete, achieving decoupling of translational and rotational components. Subsequently, the spatially variable phase error can be compensated according to the pure turntable model to obtain good microwave imaging results.

[0040] Figure 1 This is a schematic flowchart illustrating a space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation, provided in an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0041] In step S101, microwave imaging echo data after pulse compression is acquired.

[0042] The echo data includes at least the distance history, which includes at least translational and rotational motion components.

[0043] In step S102, preliminary compensation is performed on the translational motion component.

[0044] In step S103, the translational linear phase component is optimized and compensated.

[0045] In step S104, wedge transformation is used to compensate for the spatial displacement cell migration caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component.

[0046] In step S105, the echo data after the migration compensation of the rotational space-crossing distance unit is compensated again for the translational residual motion component to obtain the echo data after the translational component and the migration of the rotational space-crossing distance unit are fully compensated.

[0047] In step S106, spatial phase error compensation and azimuth compression are performed on the echo data after the translational component and the rotational spatial displacement unit migration are fully compensated to obtain the imaging result of the space target.

[0048] In some embodiments of this application, the method may be executed by a server or by a terminal device with certain processing capabilities.

[0049] In some embodiments of this application, microwave imaging echo data after pulse compression can be acquired first. Here, microwave imaging echo data after pulse compression refers to microwave imaging echo data after pulse compression.

[0050] In some embodiments of this application, preliminary compensation can be performed on the translational motion component in the microwave imaging echo data after pulse compression.

[0051] The preliminary compensation for the translational motion component may include: using the adjacent cross-correlation method to align the envelope of the echo data and determining the first-order translational motion parameters of the translational motion component based on the envelope alignment result; using the self-focusing method to estimate the global phase error and using the obtained global phase error to compensate for the azimuth data of the echo data.

[0052] The presence of translational motion components in the phase data causes the target to defocus. Since translational motion is a consistent motion across all scattering points within the target, the phase error reflected in the echo is globally consistent; that is, the compensated phase history is uniform for each range cell. Therefore, the autofocus method can be used to estimate the global phase error in the echo data, and then the estimated global phase error can be used to compensate for the azimuth data of the echo data.

[0053] The self-focusing method can be the phase gradient self-focusing method, the minimum entropy method, or other self-focusing methods; there are no restrictions here.

[0054] In some embodiments of this application, the translational motion component can also be optimized and compensated. This optimization and compensation may include: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining Keystone transform to optimize the first-order translational motion parameters, obtaining optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational linear phase component, thereby obtaining echo data with optimized and compensated translational linear phase components.

[0055] In some embodiments of this application, the Keystone transform can be used to compensate for the MTRC caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component.

[0056] Furthermore, the echo data after rotational MTRC compensation can be further compensated for with translational residual motion components to obtain echo data after complete compensation of translational components and rotational MTRC. This complete compensation of translational components and rotational MTRC can be essentially complete, meaning the echo data obtained at this point is data after almost complete compensation of translational motion components, translational linear phase components, and rotational MTRC.

[0057] Finally, spatial phase error compensation and azimuth compression can be performed on the echo data after full compensation of translational and rotational MTRC components to obtain the imaging results of the space target.

[0058] According to the technical solution provided in the embodiments of this application, the translational motion component of the microwave imaging echo data after pulse compression is initially compensated using the adjacent cross-correlation method and the self-focusing method. Then, the first-order translational motion parameter fitted by the former is used as the initial value, and the minimum image entropy of the echo data is used as the cost function. The translational linear phase component is optimized and compensated using the Keystone transform. Next, the Keystone transform is used to compensate for the spatially varied distance cell migration caused by the rotational motion component. Then, the translational residual motion component is compensated again to obtain the echo data after the translational component and the rotational MTRC are basically fully compensated. The spatially varied phase error compensation and azimuth compression are performed on the echo data after the envelope is aligned to obtain the imaging result of the space target. This achieves both efficiency and accuracy in parameter estimation, avoids the geometric distortion caused by the insensitivity of the first-order coefficients in the traditional translational phase compensation method in subsequent compensation, and effectively improves the imaging quality.

[0059] Taking microwave imaging as an example of ISAR, the transmitted signal in ISAR imaging is generally a linear frequency modulated (LFM) signal. Assume that during imaging, there is a scattering point P on the target... , The distance between the radar and the radar is , and Let P be the point. , The two-dimensional coordinate components of the transmitted signal; the length and bandwidth of the transmitted signal are respectively... and Then the echo data after pulse compression can be expressed as:

[0060] ;in, It is point P ( , The scattering coefficient of ) and These represent distance in fast time and direction in slow time, respectively; distance journey. It can be divided into translational motion. With rotational motion , , For any variable, For radar wavelength, At the speed of light, It is the symbol for imaginary numbers.

[0061] In microwave imaging of space-based targets, translational motion can be modeled up to the fourth order, and rotational motion can generally be modeled up to the second order, i.e.: ; ;in, The initial distance, For speed, For acceleration, To accelerate, To add accelerator, The relative rotational angular velocity, This is the relative rotational angular acceleration.

[0062] In the ISAR imaging process, compensating for the range offset introduced by target translation is a key step, a process often referred to as envelope alignment.

[0063] In some embodiments of this application, envelope alignment of echo data using the adjacent cross-correlation method may include:

[0064] First, the echo data of the (i-1)th pulse is converted to a high-resolution time-domain envelope; i is a positive integer greater than 1 and less than or equal to P, where P is the total number of pulses in the echo data.

[0065] Then, starting from i=2, iteratively execute the following steps until i=P: using the amplitude of the (i-1)th pulse envelope as a reference template, calculate the cross-correlation function of the i-th pulse envelope and the (i-1)th pulse envelope in the frequency domain; perform an inverse transform on the cross-correlation function to obtain a high-resolution discrete correlation curve; locate the coarse peak on the discrete correlation curve, and use the parabolic fitting method to perform sub-pixel-level precise estimation of the peak position to obtain a fractional offset; perform linear phase compensation on the i-th pulse envelope based on the offset to align each pulse envelope through time-domain shifting.

[0066] In this context, a resolution at the centimeter level is considered high resolution. In one example, a resolution greater than 3 centimeters can be defined as high resolution, or other centimeter-level numbers can be set; there are no restrictions here.

[0067] Locating the coarse peak on a discrete correlation curve can be done by using the max function on the discrete correlation curve to locate the position of the coarse peak.

[0068] Since the echo data is discretized, locating the coarse peak position can only determine the integer position of the peak. To further improve the estimation accuracy, parabolic fitting estimation is required to obtain the fractional offset.

[0069] Here, "fractional multiple" refers to an offset that is likely a fraction of the sampling interval, i.e., a sub-pixel (or sub-sampling precision) offset. In signal processing, the time shift of discrete sampled data is usually measured in integer units of sampling points. However, through interpolation methods such as cross-correlation and parabolic fitting, a more precise peak position can be estimated. This position may lie between two discrete sampling points, corresponding to a fractional multiple of the sampling interval. For example, if the estimated peak offset is 2.3 sampling points, then the "fractional multiple" is 0.3 (i.e., the decimal part). This sub-pixel estimation improves alignment accuracy and avoids residual inaccuracies caused by rounding errors.

[0070] In some implementations, the offset specifically refers to the peak offset of the i-th pulse envelope relative to the (i-1)-th pulse envelope. The peak position of the cross-correlation function reflects the relative time shift (i.e., time delay) between the two envelopes. By calculating the cross-correlation and locating the peak, it can be determined how much the i-th envelope needs to move to align with the (i-1)-th envelope. The offset can be positive or negative, representing lead or lag, respectively.

[0071] During the iteration process, the amplitude of the first envelope can be used as the initial reference template, with its offset set to zero, as the alignment reference envelope. In each subsequent iteration, the alignment reference envelope is dynamically updated. The updated alignment reference is not the envelope of the previous pulse that was just aligned, but rather calculated based on a weighted average of historical alignment results. The current pulse's envelope is cross-correlated with this weighted average alignment reference envelope to determine the relative offset. After each pulse alignment is completed, the alignment reference envelope is updated as follows: the currently aligned pulse envelope is added to the previous alignment reference envelope with a fixed weight; the previous alignment reference envelope is attenuated by an attenuation factor; the two are added together to form a new alignment reference envelope, which is used for aligning the next pulse envelope.

[0072] In other words, when using the adjacent cross-correlation method to align the envelope of the pulse compression echo, the echo data of the first pulse can be converted to a high-resolution time-domain envelope, its amplitude can be used as the initial reference template, and its offset can be recorded as zero as the alignment reference envelope.

[0073] For each subsequent pulse, the following operations are performed: The current pulse envelope and the dynamic reference template are transformed to the frequency domain, their cross-correlation function is calculated, and a high-resolution discrete correlation curve is obtained through inverse transformation; a coarse peak is located on the correlation curve, and a sub-pixel-level precise estimation of the peak position is performed using parabolic fitting to obtain a fractional offset; based on the estimated offset, corresponding linear phase compensation is applied to the current pulse data in the frequency domain to achieve accurate time-domain shifting; using the envelope of the currently aligned pulse, the reference template is dynamically updated in a first-order recursive manner to adapt to the slow changes in target scattering characteristics, enhancing the algorithm's robustness. The obtained offset is then fitted using least-squares fitting in slow time to obtain the first-order coarse estimation coefficients of the translational motion. .

[0074] After initial envelope alignment, the remaining envelope alignment error can be expressed as: ,in It can be represented as ;in, This refers to the residual constant term error in the remaining envelope alignment error. This refers to the residual velocity error within the remaining envelope alignment error. This refers to the residual acceleration error within the remaining envelope alignment error. This refers to the residual accelerometer error within the remaining envelope alignment error. This represents the residual acceleration error in the remaining envelope alignment error. Here, the residual error corresponds to the consistency cross-distance cell migration compensation error caused by the translational component.

[0075] In some embodiments of this application, a self-focusing method is used to estimate the global phase error, including: identifying k range cells with signal strength greater than a preset strength threshold in the pre-compensated echo data; k being a positive integer; identifying the point with the highest intensity in each range cell as a prominent point, truncating it with a window function, and shifting the peak position of the prominent point to zero Doppler; calculating the difference in the phase history of all prominent points and averaging it to eliminate the influence of random noise and obtain the global phase error; performing least-squares fitting on the global phase error in the azimuth direction to obtain the phase error function; removing the first-order term of the phase error function; applying the phase error function after removing the first-order term to the pulse-compressed echo data to correct the signal of each range cell; iteratively executing the above steps of determining range cells, determining prominent points, calculating the global phase error using prominent points, determining the phase error function based on the global phase error, and correcting the signal of each range cell using the phase error function, until the phase error converges or the maximum number of iterations is reached.

[0076] Signal strength refers to the energy or amplitude of a signal. Typically, strength is the square of the signal amplitude (power) or its absolute value (amplitude). High-intensity scattering points are chosen because these points have a high signal-to-noise ratio, their phase histories are less affected by noise, and they more reliably reflect the shared phase error.

[0077] In this embodiment, the value of k can be determined based on the changes in the size and resolution of the target. In one example, k can be set to a value between 20 and 50, and can be equal to 20 or 50. Alternatively, k can be set to other numerical ranges, which are not limited here.

[0078] The window function used in this embodiment is a weighted function for truncating a finite-length signal; for example, a rectangular window can be selected. In this embodiment, for each selected strong scattering point, a finite-length azimuth signal is truncated around its azimuth peak position within its range cell using the window function. The purpose of this operation is to minimize the influence of other scattering points within the range cell while capturing the main energy region of the strong scattering point as much as possible. Generally, the size of the window function gradually decreases as the number of iterations increases (the scattering points become more focused).

[0079] In the embodiments of this application, all prominent points refer to the set of prominent points of each distance cell in k distance cells.

[0080] In some embodiments of this application, determining phase error convergence can be achieved by calculating the norm (e.g., root mean square error) of the difference between the phase error vector estimated in the current iteration and the phase error vector estimated in the previous iteration. If the norm of this difference is less than a preset tolerance threshold, the phase error is considered to have converged, and the iteration stops. Meanwhile, the specific value of the maximum number of iterations can be set according to actual needs and is not limited here.

[0081] This embodiment first shifts the peak value of each prominent point to zero Doppler, which is equivalent to compensating for the linear phase component in the data at that point. The remaining data mainly contains the phase that causes defocusing. At this point, global phase error estimation allows the algorithm to focus on estimating and correcting the nonlinear phase component that causes image defocusing, thereby separating and extracting the pure phase error and ensuring the effectiveness and accuracy of the autofocusing process.

[0082] In other words, after envelope alignment, it is generally assumed that echoes from the same scattering point are located in the same range cell. At this time, the target echo still has a phase error. PGA can be used for translational phase compensation. In the image data sequence (i.e., echo data with envelope alignment achieved through preliminary compensation), select k range cells with high intensity, find the point with the highest intensity in each range cell, truncate it with a window function, and move the peak position of the prominent point to zero Doppler. Calculate the difference in the phase history of all strong scattering points and average it to eliminate the influence of random noise and obtain a global phase gradient estimate. Perform least squares fitting on the estimated phase error in the azimuth direction to truncate its first-order term. Then, apply the phase error function with the first-order term removed to the original azimuth data to correct the signal of each range cell. Repeat the above steps until the phase error converges or the maximum number of iterations is reached.

[0083] After initial phase compensation, the remaining phase compensation residual can be expressed as: ,in It can be represented as ;in, To compensate for the error in the residual constant term in the residual phase, To compensate for the residual velocity error in the residual phase, To compensate for the residual acceleration error in the residual phase, To compensate for the residual jerk error in the residual phase, This refers to the residual acceleration error in the residual phase compensation residual. Here, "residual" specifically refers to the residual global phase error compensation caused by the translational component.

[0084] In some embodiments of this application, optimizing the first-order parameters of translational motion using the Keystone transform may include:

[0085] Firstly, let's consider the first-order parameters of the translational motion. For the parameters to be optimized, establish the first compensation function. .

[0086] Then, the first compensation function is used to compensate the echo data after the initial compensation of the translational motion component, and the Keystone transform is performed on the echo data after the second compensation to obtain the signal. .

[0087] Next, the signal Calculate the image entropy to obtain ;in, Image entropy, Yes The result of discretization. It is the total energy of the image, which remains constant before and after compensation. The number of sampling points in the echo range direction. denoted as the number of azimuth sampling points, n is the corresponding distance in the discretized echo data, and m is the corresponding azimuth position in the discretized echo data.

[0088] Finally, a cost function is established based on minimizing image entropy, and the parameters are calculated. The optimal solution yields the optimized first-order parameters of the translational motion. .

[0089] In other words, the optimal translational first-order motion parameters can be searched by combining first-order phase error compensation and Keystone transformation with minimum entropy as the benchmark.

[0090] The echo data after preliminary compensation of the translational motion components can be expressed as: ;in, Let be the backscattering coefficient of the space target. Using a compensation function... Compensation signal Then, by performing the Keystone transform, the signal can be obtained. Further investigation of this signal Calculate the image entropy to obtain .

[0091] For the to be estimated Generally speaking, all other things being equal, the closer it is to the true value... The better the envelope alignment, the lower the image entropy. Therefore, a cost function can be established based on the minimum entropy, and the parameters can be calculated. The optimal solution is obtained .

[0092] In some embodiments of this application, compensating for the translational linear phase component using optimized first-order translational motion parameters may include: establishing a second compensation function using the optimized first-order translational motion parameters. The echo data after initial compensation of the translational motion components is further compensated using a dual compensation function to obtain the echo data after optimized compensation of the translational linear phase components. .

[0093] In other words, the estimated optimal parameters can be used. Establish a compensation function ,use The echo data, after initial compensation for translational motion components, is compensated again to obtain... The translational phase residual can be expressed as: ,and It can be represented as , ; The carrier frequency for transmitting signals, also known as the center frequency.

[0094] In some embodiments of this application, using Keystone transform to compensate for the MTRC caused by the rotational motion component in the echo data after translational linear phase component optimization compensation may include: converting the echo data after translational linear phase component optimization compensation from the fast time-slow time domain to the distance frequency-slow time domain to obtain the signal. , For distance frequency, For signal The slow time; through the formula For signals Perform a Keystone transformation to map it to a new slow-time coordinate. The echo data after rotation MTRC compensation was obtained. , This is the carrier frequency of the echo signal.

[0095] Among them, for the signal Performing a Keystone transformation may include: determining the signal The distance-frequency axis; for each fixed distance-frequency... Along the slow time axis Resampling is performed to obtain a new slow time series; where each coordinate in the new slow time series... All through formula Sure, For slow time axis Coordinate points in the data; based on various distance frequencies and its corresponding new slow time coordinates Determine the echo data .

[0096] In other words, the data after optimized compensation of translational linear phase components can be transformed from the fast-time-slow-time domain to the range-frequency-slow-time domain to obtain... Then, the original slow time coordinates were transformed using the Keystone transformation formula. Mapped to a new slow time coordinate superior.

[0097] The specific operation is as follows: calculate the distance-frequency axis, for each fixed distance-frequency... They all need to follow the slow time. The axis is resampled, that is, for each point in the new sequence. The slow time point corresponding to it in the original data can be calculated using the formula. Because the calculated slow time points are likely not integer indices, interpolation is needed to calculate the corresponding slow time point positions from the original data. This process is repeated for each... and After performing the above operations, a new two-dimensional signal was obtained. .

[0098] In some embodiments of this application, performing translational residual motion component compensation again on the echo data after rotational MTRC compensation may include: compensating the echo data after rotational MTRC compensation for... Taylor expansion is performed at the point; the echo data after Taylor expansion is envelope aligned again to compensate for the remaining MTRC; the echo data after the second envelope alignment is global phase error estimated again, and the global phase error obtained from the second estimation is used to compensate the azimuth data of the echo data after the second envelope alignment again, so as to obtain the echo data after the translational component and rotational MTRC are fully compensated.

[0099] In other words, the signal can exist Performing a Taylor expansion at that point yields... ;in, For the frequency modulation of the transmitted signal, , Let be the azimuth position of the i-th scattering point of the target in the target coordinate system. Let be the distance position of the i-th scattering point of the target in the target coordinate system. For rectangle functions, defined as .

[0100] As can be seen from the Taylor expansion results above, the remaining MTRC is globally consistent and can be corrected by another envelope alignment. The remaining phase error includes spatially variable and non-spatially variable errors, the latter of which can be compensated by another global autofocusing to obtain the compensated echo matrix. .

[0101] In some embodiments of this application, spatially variable phase error compensation and azimuth compression are performed on the echo data after full compensation of translational and rotational MTRC components. This may include: obtaining prominent points in the echo data after full compensation of translational and rotational MTRC components; analyzing the variation law of each scattering point in distance and azimuth using rotational modeling formula to obtain spatially variable phase error; using spatially variable phase error to compensate the echo data after full compensation of translational and rotational MTRC components to obtain echo data after spatially variable phase error compensation; and performing orientation-position fast Fourier transform (FFT) processing on the echo data after spatially variable phase error compensation to obtain the imaging result of the space target.

[0102] The method for obtaining the prominent points in the echo data after full compensation of translational and rotational MTRC components is basically the same as the method for determining the prominent points of k distance cells in the echo data after preliminary compensation, and will not be repeated here.

[0103] In some implementations, the analysis of the variation law in conjunction with the modeling formula can be carried out by first selecting prominent points to extract the phase history for coarse estimation, then establishing a cost function based on minimum entropy to optimize the parameters, and finally performing space-varying phase error compensation based on the optimized parameters.

[0104] In other words, after the second translational compensation, the echo only contains high-order spatially varying phase errors that change with the distance and azimuth of the scattering points. Since the envelope can be considered aligned at this point, the phase history of strong scattering points is extracted, and the variation of each scattering point in distance and azimuth is analyzed in combination with the modeling formula. In this way, a compensation function is constructed to compensate for the spatially varying phase errors.

[0105] In analyzing the variation patterns using modeling formulas, the rotational modeling formula shows that the rotational motion can be represented as a third-order spatial variation (changing with the scattering point position). This also causes range migration and azimuth defocusing. However, the former's compensation accuracy is lower, with the range change not exceeding one range gate; generally, only the linear part needs to be considered, as this part has already been compensated in the Keystone transform mentioned earlier. The latter, on the other hand, generally requires a residual phase compensation accuracy of less than [a certain value]. / 4, therefore, second or third order can be considered. In one example, third order can be considered.

[0106] There are generally two ways to compensate for spatially varying phase errors: block compensation and rotation parameter estimation. The latter is more commonly used under high-resolution conditions. The phase history of strong scattering points can be extracted to analyze the variation of phase history of different scattering points in distance and orientation, and the parameters can be estimated accordingly. The parameters can also be optimized using standards such as minimum entropy or contrast. The estimated parameters can then be easily compensated.

[0107] The result after compensation can be expressed as For signals By performing a Fast Fourier Transform (FFT) on the orientation, a well-focused imaging result can be obtained. ;in, For Doppler frequency, The time for synthesizing the aperture is denoted as .

[0108] from As can be seen from the expression, the two-dimensional image obtained by ISAR imaging will form a peak. The amplitude of the peak corresponds to the backscattering coefficient of the scattering point P, and the position corresponds to the coordinate position of the scattering point P (with an overall offset but unchanged shape). The ISAR image obtained by the target is the image formed by superimposing the imaging effects of all such scattering points.

[0109] The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation provided in this application may include the following steps:

[0110] The first step is to use the strong correlation of the envelopes of adjacent pulse echoes to perform envelope alignment using the adjacent cross-correlation method, and obtain the preliminary envelope alignment result. The second step is to use the phase gradient self-focusing method to estimate the global phase error, and then perform phase compensation by fitting the obtained global phase error and removing its first-order term. Both the first and second steps are translational compensation.

[0111] The third step involves using the fitted first-order translational motion parameters obtained from the envelope alignment in the first step as initial values, and then optimizing them using the Keystone transform with minimum entropy as the cost function to obtain the optimal parameter estimation results. The fourth step involves using the estimated parameters to compensate for the first-order translational phase error. The fifth step involves using the Keystone transform to compensate for the spatial variation distance cell movement caused by rotation. The sixth step involves performing another translational compensation process to compensate for the remaining translational residuals. The seventh step involves compensating for the spatial variation phase error and compressing the azimuth to obtain high-resolution imaging results.

[0112] The technical solution provided in this application utilizes the first-order motion parameters fitted by envelope alignment as initial values. Based on minimum entropy, it combines first-order translational phase compensation with KT to search for optimal parameters, balancing efficiency and accuracy in parameter estimation. The first-order phase compensation problem is improved based on the enhanced autofocus algorithm, avoiding geometric distortions in subsequent compensation caused by the insensitivity of traditional translational phase compensation methods to first-order coefficients. A translational compensation-rotational MTRC compensation-re-rotational compensation mode is selected to decouple translation and rotation, reducing the impact of rotation on translational compensation, minimizing translational residual error, and effectively improving imaging quality.

[0113] Figure 2 and Figure 3 This is a comparison chart of MTRC compensation results obtained using traditional methods and the technical solutions provided in the embodiments of this application. As the previous analysis shows, translation and rotation influence each other, and MTRC includes both uniform and spatially variable MTRC, causing errors in the traditional compensation method (envelope alignment). Figure 2 As shown. And from Figure 3 As can be seen, by using the method provided in the embodiments of this application, which involves initial translational compensation, linear phase compensation based on parameter estimation, KT compensation of spatially variable MTRC, and subsequent translational compensation of residual consistent MTRC, the MTRC is fully compensated, and both translational and rotational MTRC are well compensated.

[0114] Figure 4 and Figure 5This is a comparison diagram of imaging results obtained using traditional methods and using the technical solutions provided in the embodiments of this application. Figure 4 The imaging results are obtained by sequentially performing translational initial compensation, linear phase compensation based on parameter estimation, KT-compensated spatially variable MTRC, and spatially variable phase error compensation. No further translational compensation was performed in between to compensate for the residual consistency MTRC and phase error. Figure 4 Compared to Figure 5 However, there is defocusing due to distance and orientation. Figure 5 This is the imaging result of the spatially variable MTRC with translational initial compensation and KT compensation added. Figure 4 and Figure 5 The comparison shows that the residual terms of the first-order terms in the traditional self-focusing method have an impact on subsequent compensation.

[0115] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0116] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0117] Figure 6 This is a schematic diagram of a space-based microwave imaging translational-rotation decoupling device for space targets based on parameter estimation, provided in an embodiment of this application. Figure 6 As shown, the device includes:

[0118] The acquisition module 601 is configured to acquire microwave imaging echo data after pulse compression; the echo data includes at least the distance history, and the distance history includes at least translational motion components and rotational motion components.

[0119] The translational compensation module 602 is configured to perform preliminary compensation for the translational motion components; wherein, the preliminary compensation for the translational motion components includes: using the adjacent cross-correlation method to perform envelope alignment of the echo data, and determining the first-order translational motion parameters of the translational motion components based on the envelope alignment results; using the self-focusing method to estimate the global phase error, and using the obtained global phase error to compensate for the azimuth data of the echo data.

[0120] The translational compensation module 602 is also configured to optimize and compensate the translational motion components; wherein, the optimization and compensation includes: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining the Keystone transform to optimize the first-order translational motion parameters to obtain optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational motion components.

[0121] The rotation compensation module 603 is configured to use Keystone transformation to compensate for the space-time distance cell migration (MTRC) caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component.

[0122] The translational compensation module 602 is also configured to perform translational residual motion component compensation again on the echo data after rotational MTRC compensation, so as to obtain echo data after complete compensation of translational component and rotational MTRC.

[0123] Imaging module 604 is configured to perform spatially variable phase error compensation and azimuth compression on the echo data after full compensation of translational components and rotational MTRC, so as to obtain the imaging results of space targets.

[0124] According to the technical solution provided in the embodiments of this application, the translational motion component of the microwave imaging echo data after pulse compression is initially compensated using the adjacent cross-correlation method and the self-focusing method. Then, the first-order translational motion parameter fitted by the former is used as the initial value, and the minimum image entropy of the echo data is used as the cost function. The translational linear phase component is optimized and compensated using the Keystone transform. Next, the Keystone transform is used to compensate for the spatially varied distance cell migration caused by the rotational motion component. Then, the translational residual motion component is compensated again to obtain the echo data after the translational component and the rotational MTRC are basically fully compensated. The spatially varied phase error compensation and azimuth compression are performed on the echo data after the envelope is aligned to obtain the imaging result of the space target. This achieves both efficiency and accuracy in parameter estimation, avoids the geometric distortion caused by the insensitivity of the first-order coefficients in the traditional translational phase compensation method in subsequent compensation, and effectively improves the imaging quality.

[0125] In some implementations, the adjacent cross-correlation method is used to align the echo data envelope, including: converting the echo data of the (i-1)th pulse to a high-resolution time-domain envelope; i is a positive integer greater than 1 and less than or equal to P, where P is the total number of pulses in the echo data; iteratively executing the following steps starting from i=2 until i=P: using the amplitude of the (i-1)th pulse envelope as a reference template, calculating the cross-correlation function of the i-th pulse envelope and the (i-1)th pulse envelope in the frequency domain; performing an inverse transform on the cross-correlation function to obtain a high-resolution discrete correlation curve; locating the coarse peak on the discrete correlation curve, and using a parabolic fitting method to perform sub-pixel-level precise estimation of the peak position to obtain a fractional offset; performing linear phase compensation on the i-th pulse envelope based on the offset to align each pulse envelope through time-domain shifting.

[0126] In some implementations, a self-focusing method is used to estimate the global phase error, including: identifying k range cells with signal strength greater than a preset strength threshold in the pre-compensated echo data; k being a positive integer; identifying the point with the highest intensity in each range cell as a prominent point, truncating it with a window function, and shifting the peak position of the prominent point to zero Doppler; calculating the difference in the phase history of all prominent points and averaging it to eliminate the influence of random noise and obtain the global phase error; performing least-squares fitting on the global phase error in the azimuth direction to obtain the phase error function; removing the first-order term of the phase error function; applying the phase error function after removing the first-order term to the pulse-compressed echo data to correct the signal of each range cell; iteratively executing the above steps of determining range cells, determining prominent points, calculating the global phase error using prominent points, determining the phase error function based on the global phase error, and correcting the signal of each range cell using the phase error function, until the phase error converges or the maximum number of iterations is reached.

[0127] In some implementations, the Keystone transform is used to optimize the first-order parameters of the translational motion, including: using the first-order parameters of the translational motion... For the parameters to be optimized, establish the first compensation function. ;in, For the azimuth of the echo data, slow time, For radar wavelength, The symbol is imaginary. The echo data, after initial compensation of the translational motion components, is further compensated using the first compensation function. A Keystone transform is then performed on the recompensated echo data to obtain the signal. ;in, For the distance-time of the echo data; for the signal Calculate the image entropy to obtain ;in, Image entropy, Yes The result of discretization. It is the total energy of the image, which remains constant before and after compensation. The number of sampling points in the echo range direction. Let n be the number of azimuth sampling points, n be the corresponding distance in the discretized echo data, and m be the corresponding azimuth position in the discretized echo data; a cost function is established based on minimizing image entropy to calculate the parameters. The optimal solution yields the optimized first-order parameters of the translational motion. .

[0128] In some implementations, the translational linear phase component is compensated using optimized first-order translational motion parameters, including: establishing a second compensation function using the optimized first-order translational motion parameters. The echo data after initial compensation of the translational motion components is further compensated using a dual compensation function to obtain the echo data after optimized compensation of the translational linear phase components. .

[0129] In some implementations, the Keystone transform is used to compensate for the MTRC caused by the rotational motion component in the echo data after translational linear phase component optimization and compensation. This includes: converting the echo data after translational linear phase component optimization and compensation from the fast time-slow time domain to the range frequency-slow time domain to obtain the signal. , For distance frequency, For signal The slow time; through the formula For signals Perform a Keystone transformation to map it to a new slow-time coordinate. The echo data after rotation MTRC compensation was obtained. , The carrier frequency of the echo signal; where, for the signal Performing the Keystone transformation includes: determining the signal The distance-frequency axis; for each fixed distance-frequency... Along the slow time axis Resampling is performed to obtain a new slow time series; where each coordinate in the new slow time series... All through formula Sure, For slow time axis Coordinate points in the data; based on various distance frequencies and its corresponding new slow time coordinates Determine echo data .

[0130] In some implementations, the echo data after rotational MTRC compensation is further compensated for translational residual motion components, including: compensating the echo data after rotational MTRC compensation for translational residual motion components again. Taylor expansion is performed at the point; the echo data after Taylor expansion is envelope aligned again to compensate for the remaining MTRC; the echo data after the second envelope alignment is global phase error estimated again, and the global phase error obtained from the second estimation is used to compensate the azimuth data of the echo data after the second envelope alignment again, so as to obtain the echo data after the translational component and rotational MTRC are fully compensated.

[0131] In some implementations, spatially varied phase error compensation and azimuth compression are performed on the echo data after full compensation of translational and rotational MTRC components. This includes: acquiring prominent points in the echo data after full compensation of translational and rotational MTRC components; analyzing the variation law of each scattering point in distance and azimuth using rotational modeling formulas to obtain spatially varied phase errors; using spatially varied phase errors to compensate the echo data after full compensation of translational and rotational MTRC components to obtain echo data after spatially varied phase error compensation; and performing orientation-position Fast Fourier Transform (FFT) processing on the echo data after spatially varied phase error compensation to obtain the imaging result of the space target.

[0132] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0133] Figure 7 This is a schematic diagram of the electronic device provided in an embodiment of this application. For example... Figure 7 As shown, the electronic device 7 of this embodiment includes a processor 701, a memory 702, and a computer program 703 stored in the memory 702 and executable on the processor 701. When the processor 701 executes the computer program 703, it implements the steps in the various method embodiments described above. Alternatively, when the processor 701 executes the computer program 703, it implements the functions of each module / unit in the various device embodiments described above.

[0134] Electronic device 7 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 7 may include, but is not limited to, processor 701 and memory 702. Those skilled in the art will understand that... Figure 7 This is merely an example of electronic device 7 and does not constitute a limitation on electronic device 7. It may include more or fewer components than shown, or different components.

[0135] The processor 701 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0136] The memory 702 can be an internal storage unit of the electronic device 7, such as a hard disk or RAM of the electronic device 7. The memory 702 can also be an external storage device of the electronic device 7, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 7. The memory 702 can also include both internal and external storage units of the electronic device 7. The memory 702 is used to store computer programs and other programs and data required by the electronic device.

[0137] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0138] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it 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 can also be implemented by a computer program instructing related hardware. 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. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium may include: any entity or device capable of carrying 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), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0139] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for decoupling translational and rotational motions in space-based microwave imaging of space targets based on parameter estimation, characterized in that, include: Acquire microwave imaging echo data after pulse compression; The echo data includes at least a distance history, which includes at least translational motion components and rotational motion components. The translational motion component is initially compensated; wherein, the initial compensation of the translational motion component includes: using the adjacent cross-correlation method to perform envelope alignment on the echo data, and determining the first-order translational motion parameter of the translational motion component based on the envelope alignment result; using the self-focusing method to estimate the global phase error, and using the obtained global phase error to compensate the azimuth data of the echo data; The translational motion components are optimized and compensated; wherein, the optimization and compensation includes: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining the Keystone transform to optimize the first-order translational motion parameters to obtain optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational linear phase components. The Keystone transform is used to compensate for the spatial displacement MTRC caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component. The echo data after rotational MTRC compensation is then compensated again for translational residual motion components to obtain echo data after complete compensation of translational components and rotational MTRC. The echo data after complete compensation of translational and rotational MTRC components are subjected to spatially variable phase error compensation and azimuth compression to obtain the imaging results of the space target.

2. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 1, characterized in that, Envelope alignment of echo data is performed using the adjacent cross-correlation method, including: The echo data of the (i-1)th pulse is converted to a high-resolution time-domain envelope; i is a positive integer greater than 1 and less than or equal to P, where P is the total number of pulses in the echo data; Starting from i=2, iteratively execute the following steps until i=P: Using the amplitude of the (i-1)th pulse envelope as a reference template, calculate the cross-correlation function between the i-th pulse envelope and the (i-1)th pulse envelope in the frequency domain; The cross-correlation function is inversely transformed to obtain a high-resolution discrete correlation curve; The coarse peak is located on the discrete correlation curve, and the peak position is precisely estimated at the sub-pixel level using the parabolic fitting method to obtain the offset by a fraction of a factor. Linear phase compensation is performed on the i-th pulse envelope based on the offset to align each pulse envelope through time-domain shifting.

3. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 1, characterized in that, The global phase error is estimated using the self-focusing method, including: In the pre-compensated echo data, identify k distance cells with signal strength greater than a preset strength threshold; k is a positive integer. The point with the highest intensity in each distance cell is identified as the salient point. A window function is used to extract the salient point and the peak position of the salient point is moved to the zero Doppler position. The phase history of all prominent points is calculated and averaged to eliminate the influence of random noise and obtain the global phase error. The global phase error is fitted with least squares in the azimuth direction to obtain the phase error function; Remove the first-order term of the phase error function; The phase error function after removing the first-order term is applied to the pulse-compressed echo data to correct the signal of each range cell. The steps of determining the distance cell, determining the prominent point, calculating the global phase error using the prominent point, determining the phase error function based on the global phase error, and correcting the signal of each distance cell using the phase error function are executed iteratively until the phase error converges or the maximum number of iterations is reached.

4. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 1, characterized in that, The first-order parameters of the translational motion are optimized using the Keystone transform, including: Using the first-order parameters of the translational motion For the parameters to be optimized, establish the first compensation function. ;in, For the azimuth of the echo data, slow time, For radar wavelength, It is the symbol for imaginary numbers; The echo data after initial compensation of translational motion components is compensated again using the first compensation function, and the Keystone transform is performed on the recompensated echo data to obtain the signal. ;in, For distance-oriented fast time of echo data; For signals Calculate the image entropy to obtain ;in, Image entropy, Yes The result of discretization. It is the total energy of the image, which remains constant before and after compensation. The number of sampling points in the echo range direction. Where n is the number of azimuth sampling points, n is the distance corresponding to the discretized echo data, and m is the azimuth position corresponding to the discretized echo data. A cost function is established based on minimizing the image entropy to calculate the parameters. The optimal solution yields the optimized first-order parameters of the translational motion. .

5. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 4, characterized in that, The compensation for the translational linear phase component is performed using the optimized first-order parameters of the translational motion, including: A second compensation function is established based on the optimized first-order parameters of the translational motion. ; The second compensation function is used to further compensate the echo data after the initial compensation of the translational motion component, resulting in echo data after optimized compensation of the translational linear phase component. .

6. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 1, characterized in that, The Keystone transform is used to compensate for the MTRC caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component, including: The echo data after translational linear phase component optimization and compensation is transformed from the fast-time-slow-time domain to the range-frequency-slow-time domain to obtain the signal. , For distance frequency, For signal Slow time; Through formula For signals Perform a Keystone transformation to map it to a new slow-time coordinate. The echo data after rotation MTRC compensation was obtained. , The carrier frequency of the echo signal; Among them, for the signal Perform Keystone transformation, including: Determine signal The distance frequency axis; For each fixed distance frequency Along the slow time axis Resampling is performed to obtain a new slow time series; where each coordinate in the new slow time series... All through formula Sure, For slow time axis The coordinates of the points in the middle; Based on each distance frequency and its corresponding new slow time coordinates Determine the echo data .

7. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 6, characterized in that, The echo data after rotational MTRC compensation is then compensated again for translational residual motion components, including: The echo data after rotation MTRC compensation is in Perform Taylor expansion at this point; The echo data after Taylor expansion is envelope aligned again to compensate for the remaining MTRC; The echo data after re-envelope alignment is subjected to global phase error estimation again, and the azimuth data of the re-envelope aligned echo data is compensated again using the re-estimated global phase error, so as to obtain echo data after complete compensation of translational component and rotational MTRC.

8. The space target space-based microwave imaging translational-rotation decoupling method based on parameter estimation according to claim 1, characterized in that, The echo data after full compensation of the translational and rotational MTRC components are subjected to spatially variable phase error compensation and azimuth compression, including: Obtain the prominent points in the echo data after the translational component and rotational MTRC are fully compensated; By combining the rotation modeling formula, the variation law of each scattering point in distance and orientation is analyzed to obtain the spatially varied phase error; The echo data after complete compensation of translational and rotational MTRC components is compensated using the aforementioned spatially variable phase error to obtain spatially variable phase error compensated echo data. The echo data after spatial phase error compensation is processed by Fast Fourier Transform (FFT) of orientation position to obtain the imaging result of the space target.

9. A space-based microwave imaging translational-rotation decoupling device for space targets based on parameter estimation, characterized in that, include: The acquisition module is configured to acquire microwave imaging echo data after pulse compression. The echo data includes at least a distance history, which includes at least translational motion components and rotational motion components. The translational compensation module is configured to perform preliminary compensation on the translational motion components; wherein, the preliminary compensation on the translational motion components includes: using the adjacent cross-correlation method to perform envelope alignment on the echo data, and determining the first-order translational motion parameters of the translational motion components based on the envelope alignment results; using the self-focusing method to estimate the global phase error, and using the obtained global phase error to compensate for the azimuth data of the echo data; The translational compensation module is further configured to optimize and compensate the translational motion components; wherein, the optimization and compensation includes: using the first-order translational motion parameters as initial values, using the minimum entropy of the echo data image as the cost function, and combining the Keystone transform to optimize the first-order translational motion parameters to obtain optimized first-order translational motion parameters, and using the optimized first-order translational motion parameters to compensate the translational linear phase components. The rotation compensation module is configured to use Keystone transform to compensate for the space-time distance cell migration (MTRC) caused by the rotational motion component in the echo data after optimized compensation of the translational linear phase component. The translational compensation module is also configured to perform translational residual motion component compensation again on the echo data after rotational MTRC compensation, so as to obtain echo data after complete compensation of translational component and rotational MTRC. The imaging module is configured to perform spatially variable phase error compensation and azimuth compression on the echo data after the translational component and rotational MTRC are fully compensated, so as to obtain the imaging result of the space target.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the space target space-based microwave imaging translational rotation decoupling method based on parameter estimation as described in any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the space target space-based microwave imaging translational rotation decoupling method based on parameter estimation as described in any one of claims 1 to 8.

Images