Satellite-borne radar space target imaging method and device based on high-order motion estimation

Abstract

The invention provides a spaceborne radar space target imaging method and device based on high-order motion estimation, and belongs to the technical field of spaceborne synthetic aperture radar space target imaging, and the method comprises the steps: constructing an imaging signal model of a spaceborne radar for a space target; a time-frequency ridge line is extracted in the defocusing area, and Doppler parameters are obtained through least square fitting; on the basis of the obtained Doppler parameters, according to a mapping relation between Doppler frequencies and motion parameters, utilizing a Levenberg-Marquardt algorithm to estimate rotation parameters, and obtaining motion characteristics of the target; according to the imaging signal model and the estimated rotation parameters, high-order range migration correction and high-order space-variant phase compensation are carried out, and a focused space target spaceborne radar imaging result is obtained through two-dimensional Fourier transform. According to the method, the calculation amount is effectively reduced, the parameter estimation efficiency is improved, and the focused space target spaceborne radar image is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of spaceborne synthetic aperture radar space target imaging, and in particular to a spaceborne radar space target imaging method and device based on high-order motion estimation. Background Art

[0002] A spaceborne SAR (Synthetic Aperture Radar) space target imaging system is a surveillance imaging technology based on space-based platforms such as satellites or space shuttles, which can effectively overcome the limitation of atmospheric loss on the imaging distance and achieve high-resolution imaging through a large observation angle. However, the key technical problem faced by this system is how to develop high-resolution and low-complexity imaging algorithms. Since both the radar and the target are in high-speed orbital motion, although this motion helps to improve the azimuth resolution, it also introduces range migration and spatially variant phase errors, resulting in imaging defocus. Especially when the target rotates relatively uniformly and accelerates, high-order spatially variant errors will occur, which are difficult to solve by traditional methods.

[0003] For the above problems, the existing technologies mainly adopt two categories: range migration correction and spatially variant phase error compensation. Although the Keystone Transform (KT) and its generalized forms can correct part of the range migration, they cannot handle spatially variant migration, and their effects will decline as the imaging time increases and the bandwidth improves. In terms of spatially variant phase error compensation, the signal domain correlation method has a large amount of calculation and is prone to introducing false targets; the image domain optimization method has high robustness, but is greatly affected by the initial value deviation and has a high computational complexity. Therefore, the existing methods have problems of insufficient estimation accuracy and large computational amount in high-order range migration correction and high-order spatially variant phase error compensation, and there is an urgent need to develop an efficient algorithm suitable for spaceborne SAR imaging. Summary of the Invention

[0004] To solve the problems of range migration and spatially variant phase error in the spaceborne radar space target imaging, the present invention provides a spaceborne radar space target imaging method and device based on high-order motion estimation. By establishing a signal model including high-order motion, a high-order motion parameter estimation model based on time-frequency ridge extraction is proposed, and the Doppler parameters of the time-frequency ridge in the defocused area are used to solve the optimal rotation parameters in combination with the Levenberg-Marquardt algorithm (LMA). In addition, the present invention proposes an improved second-order scaling operation and a third-order phase compensation function to correct high-order spatially variant range migration and compensate high-order spatially variant phase errors, so as to achieve high-resolution image focusing. The present invention also uses the Levenberg-Marquardt algorithm (LMA) to estimate motion parameters in the signal domain, effectively reducing the computational amount and improving the parameter estimation efficiency.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A spaceborne radar space target imaging method based on high-order motion estimation, which performs high-order motion estimation to compensate for high-order spatially variant range migration error and high-order spatially variant phase error, includes the following steps:

[0007] Step 1: Construct an imaging signal model of the spaceborne radar for the space target, considering the relative motion characteristics of the radar and the target to describe the characteristics of the echo signal;

[0008] Step 2: Extract the time-frequency ridge line in the defocused area and obtain the Doppler parameters through least squares fitting;

[0009] Step 3: Based on the obtained Doppler parameters, according to the mapping relationship between the Doppler frequency and the motion parameters, use the Levenberg-Marquardt algorithm to estimate the rotation parameters and obtain the motion characteristics of the target;

[0010] Step 4: According to the imaging signal model and the estimated rotation parameters, perform high-order range migration correction and high-order spatially variant phase compensation, and through two-dimensional Fourier transform, obtain the focused spaceborne radar imaging result of the space target.

[0011] The present invention also provides a spaceborne radar space target imaging device based on high-order motion estimation, including the following modules:

[0012] An imaging signal model construction module, which constructs an imaging signal model of the spaceborne radar for the space target, considering the relative motion characteristics of the radar and the target to describe the characteristics of the echo signal;

[0013] A fitting module, which extracts the time-frequency ridge line in the defocused area and obtains the Doppler parameters through least squares fitting;

[0014] A motion characteristic acquisition module, which based on the obtained Doppler parameters, according to the mapping relationship between the Doppler frequency and the motion parameters, uses the Levenberg-Marquardt algorithm to estimate the rotation parameters and obtain the motion characteristics of the target;

[0015] A result acquisition module, which according to the estimated rotation parameters, performs high-order range migration correction and high-order spatially variant phase compensation, and through two-dimensional Fourier transform, obtains the focused spaceborne radar imaging result of the space target.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps of the above-mentioned spaceborne radar space target imaging method based on high-order motion estimation.

[0017] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-mentioned spaceborne radar space target imaging method based on high-order motion estimation.

[0018] Beneficial effects:

[0019] By introducing time-frequency ridge extraction and rotation parameter estimation techniques, the present invention breaks through the limitations of traditional signal domain correlation methods and image domain optimization methods. In terms of parameter estimation, it can accurately obtain the parameters of the relative rotation of the target, providing a reliable basis for subsequent processing. Further, the second-order scaling operation proposed by the present invention can effectively correct the second-order space-variant range migration, while the third-order phase compensation function can compensate for both second-order and third-order space-variant phase errors, thus significantly improving the imaging accuracy. In addition, by avoiding high-order moment operations and a large number of fast Fourier transform (FFT) operations, the present invention greatly reduces the algorithm complexity and improves the calculation efficiency. Brief description of the drawings

[0020] Figure 1 is a flowchart of a spaceborne radar space target imaging method based on high-order motion estimation according to the present invention;

[0021] Figure 2 is a schematic diagram of a spaceborne radar space target imaging device based on high-order motion estimation according to the present invention;

[0022] Figure 3 is a schematic diagram of the target satellite model for simulation;

[0023] Figure 4a , Figure 4b , Figure 4c is a comparison diagram of imaging results; among them, Figure 4a is the imaging result of the RD (range-Doppler) algorithm, Figure 4b is the imaging result of the RDK (range-Doppler based on Keystone transform (KT)) algorithm, Figure 4c is the imaging result of the method proposed by the present invention. Detailed implementation manners

[0024] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] As Figure 1 shown, according to an embodiment of the present invention, a spaceborne radar space target imaging method based on high-order motion estimation includes the following steps:

[0026] Step 101: Establish an imaging signal model of the spaceborne radar for the space target, including:

[0027] Assume that the radar uses a linear frequency modulation signal as the transmitted signal. The echo from the target is down-converted to the baseband signal, and then range pulse compression is performed. The range-frequency domain signal after translational compensation can be expressed as:

[0028] (1)

[0029] where, is the range frequency, is the slow time, p represents the scattering point index value, is the total number of scattering points, is a complex amplitude, is the carrier frequency, is the speed of light, represents a scattering point on the space target is the instantaneous slant range introduced by the relative rotation, represents the exponential function, and j represents the imaginary unit.

[0030] Due to the characteristics of the relative uniform accelerated rotational motion of the space target, the third-order Taylor expansion of the trigonometric function is used to accurately represent the instantaneous slant range introduced by the relative rotation, as shown below:

[0031] (2)

[0032] where, represents the scattering point 's initial coordinates, is the relative rotational speed, is the relative rotational acceleration, represents at the moment of the rotation angle.

[0033] Next, KT (Keystone Transform) is used for linear range migration correction. The range-frequency domain resampling used by KT can be expressed as:

[0034] (3)

[0035] where, is the resampled slow time, , N , , , , ,

[0037] , ,

[0030] , ,

[0032] ,

[0036] ,

[0035] , , , , ,

[0033] , , , , , , , ,

[0034] ,

[0031] , , , , , , ,

[0038] , , m , , , represents the number of slow time samples, and m represents the time coordinate.

[0036] Then, the range-frequency domain signal after KT can be expressed as:

[0037] (4)

[0038] where, represents the signal wavelength, and p represents the scattering point index value.

[0039] Step 102: Perform time-frequency ridge extraction in the defocused area to obtain Doppler parameters, including:

[0040] Based on the established imaging signal model, use the image domain block technology to extract the equivalent echo data in the defocused area. This echo data can be considered as the space-invariant equivalent echo and can be expressed as:

[0041] (5)

[0042] where, is the azimuth echo data of the th range cell, represents the number of scatterers included in the nth range cell, n represents the range cell serial number, represents the range resolution, and p represents the scatterer index value.

[0043] Subsequently, adopt the WVD (Wigner-Ville Distribution) as the time-frequency analysis method to obtain the high-resolution time-frequency distribution of the selected range gate, which can be expressed as:

[0044] (6)

[0045] where, is the time-frequency distribution, is the frequency coordinate, is the time coordinate, and the superscript represents the conjugate operation, represents the delay time, and f represents the frequency.

[0046] Then, through the time-frequency ridge extraction technology, the time-frequency ridge composed of the instantaneous frequencies of the scatterers can be obtained , and the time-frequency ridge extraction can be expressed as:

[0047] (7)

[0048] where, represents the k value when finding the maximum value of the given function.

[0049] Subsequently, through the least squares fitting, the Doppler parameters corresponding to the time-frequency ridge can be obtained, as follows:

[0050] (8)

[0051] where, the intermediate parameter , the intermediate parameter , is the vector composed of the Doppler parameter estimation values, and the superscript T represents the transpose of the matrix, Denote the first-order coefficient, second-order coefficient, and third-order coefficient obtained by fitting the Doppler frequency, i.e., the Doppler parameter estimation values.

[0052] Step 103: Estimate the rotation parameters based on the obtained Doppler parameters, including:

[0053] According to Equation (5) and the imaging signal model, the Doppler frequency of the scatterer can be expressed as:

[0054] (9)

[0055] where, is the time-frequency ridge line obtained after extraction, is the Doppler instantaneous frequency, is the linear component of the Doppler frequency, is the quadratic component of the Doppler frequency. Denote taking the th range gate azimuth echo data phase.

[0056] Furthermore, the equivalent matrix form of the above equation can be expressed as:

[0057] (10)

[0058] where, A is the vector composed of the true values of the Doppler parameters, i.e., A is the true value; is the position of the scatterer, is the rotational motion mapping matrix, as shown below:

[0059] (11)

[0060] It can be seen from this that the Doppler parameters of the time-frequency ridge line are related to the rotational parameters of the target, and the unknown rotational parameters can be represented by a three-element parameter vector , and the following uses a non-linear regression model to solve the rotational parameters. The constructed non-linear regression model for estimation is as follows:

[0061] (12)

[0062] where, is the estimated value, , are respectively the estimated values, represents solving the independent variable value x that minimizes the objective function, represents the square of the two-norm.

[0063] The present invention uses the Levenberg-Marquardt algorithm based on the non-linear regression method for solution, and can efficiently obtain the estimation result of the rotation parameters. The Levenberg-Marquardt algorithm constructs a weighted linear approximation problem in each iteration, and introduces a damping factor to adaptively adjust the step size between gradient descent and Gauss-Newton update, so as to gradually approach the solution that minimizes the sum of squared residuals.

[0064] Step 104: Use the imaging signal model in Step 101 and the rotation parameters in Step 103 to perform high-order range migration correction and high-order space-variant phase compensation, including:

[0065] According to Equation (4), the equivalent frequency-domain signal after block division can be expressed as:

[0066] (13)

[0067] where represents the block number.

[0068] Furthermore, according to the rotation parameters, the improved second-order scaling operation proposed by the present invention is as follows:

[0069] (14)

[0070] where represents the range frequency, the range frequency after scaling, is the serial number of the range cell where the scatterer P is located.

[0071] Substituting Equation (14) into Equation (13), we can obtain:

[0072] (15)

[0073] where the high-order space-variant range migration has been compensated.

[0074] Furthermore, a space-variant phase error compensation function can be constructed as follows:

[0075] (16)

[0076] After performing high-order space-variant phase error compensation, the focused image result can be obtained through FFT (Fast Fourier Transform)

[0077] (17)

[0078] where 、 respectively represent performing Fourier transforms on the time dimension and the range dimension.

[0079] Therefore, the present invention can compensate for the second-order space-variant range migration error and the second- and third-order space-variant phase errors, and obtain high-resolution space target radar imaging results.

[0080] Embodiment:

[0081] This embodiment is based on the orbital parameters of the Chinese space station and LT-1. Taking the Chinese space station as the monitoring satellite and LT-1 as the target satellite for imaging, the results obtained by applying the proposed method for imaging are calculated, and the feasibility of the proposed method is verified by comparison. The model of the target satellite is as Figure 3 shown. O-XYZ is the coordinate system where the scatter point is located. X represents the azimuth axis, Y represents the range axis, and Z represents the altitude axis. The radar parameters used in the simulation are shown in Table 1.

[0082] Table 1 Radar Parameters of the Monitoring Satellite

[0083] The imaging results obtained by using three methods are as Figure 4a , Figure 4b , Figure 4c shown. Figure 4a is the imaging result of the RD (Range-Doppler algorithm) algorithm. Figure 4b is the imaging result of the RDK (Depth Perception algorithm) algorithm. Figure 4c is the imaging result of the method proposed by the present invention.

[0084] The image contrast (IC) and image entropy (IE) can reflect the performance of the imaging algorithm and are widely used as quantitative indicators of imaging quality. An imaging algorithm with a larger IC and a smaller IE has better focusing performance. The IC and IE of the above three results are compared below, as shown in Table 2.

[0085] Table 2 Comparison Results of Imaging Performance

[0086] It can be observed that the method proposed by the present invention has the highest IC and the lowest IE, which proves the effectiveness of the proposed method. This verifies the superiority of the proposed method and proves its potential for application in future space target SAR imaging tasks.

[0087] As Figure 2 shown, the present invention also provides a space target imaging device for spaceborne radar based on high-order motion estimation, including the following modules:

[0088] An imaging signal model construction module constructs an imaging signal model of a spaceborne radar for a space target, taking into account the relative motion characteristics between the radar and the target to describe the characteristics of the echo signal;

[0089] A fitting module extracts the time-frequency ridge line in the defocused area and obtains the Doppler parameters through least squares fitting;

[0090] A motion feature acquisition module, based on the obtained Doppler parameters, according to the mapping relationship between the Doppler frequency and the motion parameters, uses the Levenberg-Marquardt algorithm to estimate the rotation parameters and obtain the motion characteristics of the target;

[0091] A result acquisition module, according to the estimated rotation parameters, performs high-order range migration correction and high-order space-variant phase compensation, and through two-dimensional Fourier transform, obtains the focused spaceborne radar imaging result of the space target.

[0092] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the above-mentioned spaceborne radar space target imaging method based on high-order motion estimation are implemented.

[0093] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned spaceborne radar space target imaging method based on high-order motion estimation are implemented.

[0094] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented in various computer languages, for example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.

[0095] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementation in the processFigure 1 a process or processes and / or blocks Figure 1 means for the functions specified in a block or blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions in the process Figure 1 a process or processes and / or blocks Figure 1 specified in a block or blocks.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions in the process Figure 1 a process or processes and / or blocks Figure 1 specified in a block or blocks.

[0098] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0099] It is apparent that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A spaceborne radar space target imaging method based on high-order motion estimation, characterized in that Perform high-order motion estimation to compensate for high-order spatially variant range migration error and high-order spatially variant phase error, including the following steps: Step 1: Construct an imaging signal model of the spaceborne radar for a space target, considering the relative motion characteristics between the radar and the target to describe the characteristics of the echo signal; Step 2: Extract the time-frequency ridge line in the defocused area and obtain the Doppler parameters through least squares fitting; Step 3: Based on the obtained Doppler parameters, according to the mapping relationship between the Doppler frequency and the motion parameters, use the Levenberg-Marquardt algorithm to estimate the rotation parameters and obtain the motion characteristics of the target; Step 4: According to the imaging signal model and the estimated rotation parameters, perform high-order range migration correction and high-order spatially variant phase compensation. After two-dimensional Fourier transform, obtain the focused spaceborne radar imaging result of the space target.

2. The space target imaging method for spaceborne radar based on high-order motion estimation according to claim 1, characterized in that In the above Step 1, constructing the imaging signal model of the spaceborne radar for a space target includes: assuming that the radar uses a linear frequency modulation signal as the transmitted signal, down-convert the echo from the target to the baseband signal, then perform range pulse compression and translational compensation to obtain the range-frequency domain signal, and use the third-order Taylor expansion of trigonometric functions to represent the instantaneous slant range introduced by relative rotation, establish a complete imaging signal model, consider the high-order range migration and spatially variant phase caused by relative rotation, and obtain the range-frequency domain signal after keystone transform.

3. The space target imaging method of the spaceborne radar based on high-order motion estimation according to claim 1, wherein The above Step 2 includes: using the image domain block technology to extract the equivalent echo data in the defocused area and regard it as the space-invariant equivalent echo; using the Wigner-Ville distribution as the time-frequency analysis method to obtain the high-resolution time-frequency distribution; obtaining the time-frequency ridge line composed of the instantaneous Doppler frequencies of the scatterers through the time-frequency ridge line extraction technology, and then using least squares fitting to obtain the Doppler parameters corresponding to the time-frequency ridge line.

4. The space target imaging method of the spaceborne radar based on high-order motion estimation according to claim 1, characterized in that The above Step 3 includes: establish a nonlinear regression model according to the relationship between the Doppler parameters and the rotation parameters; use the Levenberg-Marquardt algorithm to solve the nonlinear regression model to obtain the estimation result of the rotation parameters.

5. The method for spaceborne radar space target imaging based on high-order motion estimation according to claim 1, characterized in that: In the above Step 4, performing high-order range migration correction and high-order spatially variant phase compensation includes: according to the estimated rotation parameters, perform an improved second-order scaling operation to compensate for the high-order spatially variant range migration; construct a spatially variant phase error compensation function to compensate for the high-order spatially variant phase error, and finally obtain the focused image result through Fourier transform to achieve high-resolution imaging.

6. The spaceborne radar space target imaging method based on high-order motion estimation according to claim 2, wherein The distance frequency-domain signal after the cuneiform transformation is and is expressed as: （4） wherein, is the range frequency, is the slow time after resampling, is the total number of scatterers, is a complex amplitude, is the carrier frequency, is the speed of light, represents the exponential function, represents the scatterer initial coordinates, is the relative rotational speed, is the relative rotational acceleration, j represents the imaginary unit, p represents the scatterer index value, represents the signal wavelength.

7. The space target imaging method for spaceborne radar based on high-order motion estimation according to claim 4, wherein Using the Levenberg-Marquardt algorithm to solve the nonlinear regression model to obtain the estimation result of the rotation parameters includes: Obtain the vector A composed of the true values of the Doppler parameters: （10） where A is a vector composed of the true values of Doppler parameters; is the position of the scattering point, the superscript T represents the transpose of the matrix, and n represents the nth range gate, is the range resolution, is the mapping matrix, as shown below: （11） Among them, is the relative rotational speed, is the relative rotational acceleration, represents the signal wavelength; The estimation result of the rotation parameters is obtained through the following formula: （12） wherein, is the estimated value of, the unknown rotation parameter , , are respectively the estimated values of, represents solving for the value of the rotation parameter x that minimizes the objective function, represents the square of the two-norm, and the superscript T represents the transpose of the matrix, represents the vector composed of the Doppler parameter estimated values.

8. The method for spaceborne radar space target imaging based on high-order motion estimation according to claim 5, characterized in that: The improved second-order scaling operation is as follows: （14） Among them, represents the range frequency, represents the range frequency after scaling, is the serial number of the range cell where the scatterer P is located; are respectively the estimated values of, represents the scatterer initial coordinates of, is the relative rotational speed, is the relative rotational acceleration, is the range resolution, is the slow time after resampling; Constructing space-variable phase error compensation function as follows: （16）。 9. A space-borne radar space target imaging device based on high-order motion estimation, characterized in that: Perform high-order motion estimation to compensate for high-order spatially variant range migration error and high-order spatially variant phase error, including the following modules: An imaging signal model construction module that constructs an imaging signal model of the spaceborne radar for a space target, considering the relative motion characteristics between the radar and the target to describe the characteristics of the echo signal; A fitting module that extracts the time-frequency ridge line in the defocused area and obtains the Doppler parameters through least squares fitting; The motion feature acquisition module, based on the obtained Doppler parameters, according to the mapping relationship between the Doppler frequency and the motion parameters, uses the Levenberg-Marquardt algorithm to estimate the rotation parameters and acquire the motion features of the target; The result acquisition module performs high-order range migration correction and high-order space-variant phase compensation according to the estimated rotation parameters, and through two-dimensional Fourier transform, obtains the focused space target spaceborne radar imaging result.

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

11. A non-transitory computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of a spaceborne radar space target imaging method based on high-order motion estimation as described in any one of claims 1 to 8.

Images