Space-borne SAR scene-matching mode time-domain imaging method

Abstract

The application provides a spaceborne SAR scene matching mode time domain imaging method, belongs to the field of synthetic aperture radar, and mainly solves the time-varying parameter problem in spaceborne scene matching SAR time domain imaging, adopts a new method of phase compensation based on time-varying parameters, adapts to the time-varying characteristic of the spaceborne scene matching SAR mode parameters, adopts a method of accurately calculating the ground beam ellipse to irradiate the imaging grid points, avoids superimposing the energy of the imaging grid points which is not irradiated, solves the ground beam ellipse shape change problem under the spaceborne SAR scene matching mode, simultaneously avoids time domain imaging aliasing, and can realize time domain high-resolution imaging. The application has full-aperture data processing capacity, can adapt to the influence caused by the time-varying parameters, and is a high-efficiency and accurate imaging processing method suitable for spaceborne scene matching SAR ground observation requirements.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of synthetic aperture radar (SAR), and particularly relates to a spaceborne SAR scene matching mode time domain imaging method. BACKGROUND

[0002] Spaceborne synthetic aperture radar (SAR) is an active remote sensing radar system working in the microwave band, which can realize active all-weather and all-time high-resolution imaging. Synthetic aperture radar uses signal processing methods to equivalently increase the length of the azimuth antenna by beam irradiation accumulation, thereby obtaining high-resolution imaging capability in the azimuth direction. Spaceborne synthetic aperture radar can be used for ground surface monitoring under complex weather conditions without being limited by sunlight conditions and weather conditions, and plays an important role in battlefield reconnaissance, disaster monitoring and early warning, topographic mapping, resource exploration and the like.

[0003] Traditional spaceborne synthetic aperture satellites are usually operated in near-polar orbits to observe high-latitude areas. Due to the orbit, the traditional synthetic aperture radar imaging band is approximately distributed along the "north-south" direction, which leads to a certain observation angle with many hot observation areas. When the spaceborne synthetic aperture radar images a target scene at a certain angle with the satellite orbit (oblique scene), the scene is easy to deviate from the imaging band, resulting in imaging failure. To solve this problem, the traditional synthetic aperture radar uses a multi-observation band mode to completely image the oblique scene, which is specifically divided into multi-orbit revisit and single-orbit time-sharing observation. The multi-orbit revisit observation has problems such as long data acquisition time and poor consistency of multi-imaging band images, and the single-orbit time-sharing observation has problems such as poor azimuth resolution of images.

[0004] Spaceborne scene matching SAR can well solve the above problems, which directly generates an imaging band matched with the scene by adjusting the beam pointing. The spaceborne scene matching SAR mode has very good adaptability to oblique scenes and high data acquisition effectiveness. Due to the large squint configuration and two-dimensional beam scanning in this mode, the echo signal parameters have strong time-varying characteristics, and in the spaceborne SAR scene matching mode, the Doppler center changes dramatically due to the two-dimensional scanning of the radar beam, and the Doppler bandwidth is large. Due to the system configuration of the spaceborne SAR scene matching mode, the pulse repetition frequency is limited, and in the scene matching sliding spotlight mode, the Doppler spectrum aliasing occurs, thereby causing time domain imaging aliasing. SUMMARY

[0005] Therefore, the application provides a spaceborne SAR scene matching mode time domain imaging method, which can adapt to the time-varying characteristics of the spaceborne scene matching SAR mode parameters, solves the problem of the ground beam ellipse shape change in the spaceborne SAR scene matching mode, avoids time domain imaging aliasing, and realizes time domain high-resolution imaging.

[0006] To achieve the above object, the technical scheme of the application is as follows:

[0007] The spaceborne SAR scene matching mode time domain imaging method comprises the following steps:

[0008] Step 1: phase compensation is performed on the target backscattering echo after distance direction Fourier transform to obtain a signal distance spectrum; inverse distance direction Fourier transform is performed on the signal distance spectrum to obtain a distance direction compressed signal in the time domain; wherein the phase compensation is performed in the following manner: based on the second phase term of the target backscattering echo after distance direction Fourier transform, a distance direction time-varying parameter compensation and quadratic phase compensation filter F RD is generated; the target backscattering echo after distance direction Fourier transform is multiplied by F RD to obtain the signal distance spectrum;

[0009] Step 2: scene matching imaging grids are divided based on the radar beam pointing direction;

[0010] Step 3: the ground beam ellipse is calculated to perform imaging grid point irradiation judgment, and the specific process is as follows:

[0011] The corresponding azimuth time is selected, the origin of the coordinate system is moved to the position of the radar platform at the corresponding time, the coordinate system is rotated around the Z axis, and the positive direction of the Y axis is rotated to the direction in which the radar ground projection point points to the beam center; the coordinate system is rotated around the X axis, and the positive direction of the Z axis points to the ground beam center; the beam center rotation angle is determined according to the attitude information of the radar satellite, and the Z axis is rotated by the response angle; according to the two-dimensional beam width of the radar beam, an elliptical cone equation is established to perform beam irradiation target judgment on the imaging grid points, and the specific process is as follows:

[0012] If the Z value of the coordinate of the imaging grid point in the transformed coordinate system is greater than the X coordinate and the Y coordinate corresponding to the Z value on the elliptical cone, it is considered that the target is inside the elliptical cone and is in the beam irradiation range, otherwise, it is considered that the target is outside the beam irradiation range, the information about whether the imaging grid point is irradiated by the beam at each time is obtained, the calculated ground beam ellipse template is obtained, and the template size is the maximum rectangular imaging grid size covered by the beam; a certain number of position beam centers are selected to obtain the corresponding ground beam ellipse template; in the backward projection process, different ground beam ellipse templates are selected according to the beam center at different positions.

[0013] Step four, the signal after phase compensation and distance compression processing is processed by distance upsampling; then, one column of echo signals after distance pulse compression is read each time, the signal corresponding to the slant range delay of the scene matching imaging grid point is projected onto the corresponding grid point, the Doppler phase corresponding to the imaging grid point at the azimuth time is compensated at the same time, the contribution of each pulse at each azimuth time is coherently accumulated, and finally the reconstruction result of each pixel point in the imaging grid is obtained;

[0014] Step five, the reconstruction result obtained in step four is processed by phase preserving to obtain a final image.

[0015] In step two, the beam pointing is obtained based on the actual radar satellite three-axis pointing in the satellite return auxiliary data, the wave foot is calculated by combining the orbit information to obtain the wave foot trend, and the scene matching imaging grid is realized according to the wave foot trend.

[0016] In step two, the way of realizing the scene matching imaging grid according to the wave foot trend is to determine the scene observation band extension direction, and to establish the imaging grid along the scene observation band extension direction according to the resolution requirement.

[0017] In step four, the distance upsampling method is frequency domain zero padding interpolation.

[0018] In step five, the phase preserving is specifically: the Doppler phase is restored according to the scene observation band center slant range to ensure that the image phase corresponds to the corresponding physical position.

[0019] Beneficial effects:

[0020] The application is a new method of phase compensation based on time-varying parameters, which adapts to the time-varying characteristics of the spaceborne scene matching SAR mode; the method of accurately calculating the ground beam ellipse to irradiate the imaging grid point is adopted to avoid superimposing the energy not irradiated in the imaging grid point, solve the problem of ground beam ellipse shape change in the spaceborne SAR scene matching mode, and avoid time domain imaging aliasing, so that time domain high resolution imaging can be realized. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 It is the flow chart of the spaceborne SAR scene matching mode time domain imaging method of the application;

[0022] Figure 2 It is the geometric relationship of the spaceborne SAR scene matching mode of the application;

[0023] Figure 3 It is the schematic diagram of dividing the scene matching imaging grid of the application;

[0024] Figure 4 It is the schematic diagram of accurately calculating the ground beam ellipse beam irradiation of the application;

[0025] Figure 5 is the nine-point target simulation result of the satellite-borne SAR scene matching mode of the application;

[0026] Figure 6 is the simulation evaluation result of the scene center point target of the application;

[0027] Figure 7 is the surface target simulation result of the application. DETAILED DESCRIPTION

[0028] The application will be described in detail below with reference to the accompanying drawings and examples.

[0029] The flow chart of the satellite-borne SAR scene matching mode time-domain imaging method of the application is shown in Figure 1 , and the specific steps include:

[0030] Step one, phase compensation and range signal compression processing based on time-varying parameters.

[0031] In the satellite-borne scene matching SAR mode, the Cartesian coordinate system is established as follows: the origin O coincides with the center of the imaging observation band, the direction away from the center of the Earth and perpendicular to the surface direction is defined as the Z axis, the direction of the beam coverage band is defined as the Y axis, i.e. the azimuth direction, and the X axis direction is determined according to the right-hand rule from the Y axis and the Z axis, i.e. the range direction, the X axis and the Y axis represent the range and azimuth directions respectively. The satellite-borne SAR scene matching mode time-domain imaging method introduced in the application is based on the fact that the ground wave foot speed corresponding to the beam center of the satellite-borne SAR scene matching mode is constant, and the geometric relationship of the satellite-borne SAR scene matching mode of the application is shown in Figure 2 , assuming that the ground wave foot speed corresponding to the beam center of the LSTM-SAR is constant, V r is the satellite speed, H is the satellite height, β0 is the incidence angle at the sampling center time of the echo signal, R gref represents the center slant range projection on the ground surface at the sampling center time of the echo signal, R gref is defined, and the angle between the vertical direction of the beam coverage band and the scene matching observation angle is marked as α, θ is defined as the scene oblique angle, which represents the angle between the beam coverage band and the satellite speed vector, assuming that an arbitrary target P is located in the imaging observation band, with coordinates (x p , y p , 0), the instantaneous slant range between the antenna phase center and the target P is R, and the center synthetic aperture time when the beam center passes through the target P is represented as t p . The instantaneous slant range R of the target P is as follows:

[0032]

[0033] where R g and tp There is a one-to-one correspondence between the coordinates of the target and the coordinates of the target.

[0034]

[0035] If a traditional linear frequency modulated (LFM) signal is used as the transmitted signal, and the echo signal from the target P is demodulated to the fundamental frequency, the demodulated echo signal can be represented as follows:

[0036]

[0037] Where c is the speed of light, τ represents the time in the range direction (faster), t represents the time in the azimuth direction (slower), f0 is the signal carrier frequency, and K... r The signal frequency is regulated. Since the signal amplitude does not affect imaging, the signal amplitude term is omitted in (3). Due to the two-dimensional sweep of the radar beam in the spaceborne SAR scene matching mode, the beam center slant range changes greatly. Therefore, the spaceborne SAR scene matching mode adopts a sliding receiver window system and assumes that a linear frequency modulated (LFM) signal is used for transmission to obtain the backscattered echo of target P:

[0038]

[0039] Where R min (t) represents the opening time of the radar system's receiving window at each azimuth. The radar system adapts to the effects of changes in center slant range by adjusting the opening time of the receiving window. In a variable repetition frequency (VRF) system, the opening time of the receiving window is calculated from the ambiguity number corresponding to the pulse repetition frequency and the delay relationship.

[0040] Perform a distance-to-Fourier transform on (4) to obtain

[0041]

[0042] Where f τ The term in (5) represents the range frequency. The first term is the range-independent Doppler phase, the second term represents the range-directed secondary phase that should be removed, and the third term represents the range migration-related phase. Therefore, a range-directed time-varying parameter compensation and secondary phase compensation filter F can be generated based on the second phase term in (5). RD :

[0043]

[0044] Multiplying equation (5) by equation (6), we obtain the signal distance spectrum as follows:

[0045]

[0046] Performing a range-to-inverse Fourier transform on equation (7) yields the range-compressed signal in the time domain as follows:

[0047]

[0048] where B r denotes the transmitted LFM signal bandwidth, the function sinc(·) has the following definition:

[0049]

[0050] Step two, divide the scene matching imaging grid based on radar beam pointing.

[0051] In the actual work of the radar system, due to the objective non-ideal factors such as satellite body rotation error, the preset radar beam irradiation pointing may deviate from the actual radar beam pointing, so it is necessary to obtain the beam pointing based on the actual radar satellite three-axis pointing in the satellite return auxiliary data, and combine the orbit information to calculate the wave foot. According to the wave foot direction, the scene observation band extension direction is determined, and the imaging grid is established along the scene observation band extension direction according to the resolution requirement, that is, the scene matching imaging grid.

[0052] Step three, accurately calculate the ground beam ellipse to judge the irradiation of the imaging grid point.

[0053] In the scene matching mode of spaceborne SAR, due to the two-dimensional scanning of the radar beam, the Doppler center changes dramatically and the Doppler bandwidth is large. Due to the limitation of the system configuration of spaceborne SAR scene matching mode, the pulse repetition frequency is limited, and in the scene matching sliding spotlight mode, the Doppler spectrum aliasing occurs, which leads to time domain imaging aliasing.

[0054] In order to solve this problem, the method of accurately calculating the ground beam ellipse to judge the irradiation of the imaging grid point is adopted, which avoids the superposition of unirradiated energy on the imaging grid point.

[0055] The beam irradiation target determination is divided into four steps. First, select the corresponding azimuth time, and move the origin of the coordinate system to the position of the radar platform at the corresponding time. Second, rotate the coordinate system around the Z axis, and rotate the positive direction of the Y axis to the direction of the beam center pointed by the radar ground projection point. Third, rotate the coordinate system around the X axis, and point the positive direction of the Z axis to the ground beam center. Fourth, determine the beam center rotation angle according to the attitude information of the radar satellite, and rotate the Z axis by the corresponding angle. Fifth, according to the two-dimensional beam width of the radar beam, the elliptical cone equation is established to judge the beam irradiation target of the imaging grid point.

[0056] If the Z value of the imaging grid point in the coordinate system after transformation is greater than the X coordinate and Y coordinate corresponding to the Z value on the elliptical cone surface, it is considered that the target is inside the elliptical cone surface and within the beam irradiation range, otherwise it is outside the beam irradiation range. Thus, the information of whether each imaging grid point is irradiated by the beam at each moment can be obtained, and the accurate ground beam elliptical template can be obtained, and the size of the template is the maximum square imaging grid size that can be covered by the beam.

[0057] A proper number of position beam centers are selected to obtain the corresponding accurate ground beam elliptical template. In the case of less than 5% change in the number of template points, it is considered that the change of beam projection on the ground can be ignored. In the back projection process, different ground beam elliptical templates are selected according to the beam center at different positions.

[0058] Step four, back projection imaging processing.

[0059] The back projection imaging algorithm of the scene matching mode of the spaceborne SAR is to process the range direction pulse compression-azimuth time domain signal, and the core idea is to realize the final imaging by processing each pulse combined with the real position of the antenna phase center. In the imaging process, the slant range between each grid point and the radar satellite needs to be calculated, and the corresponding echo is projected onto the grid point according to the time delay corresponding to the slant range. First, the signal after phase compensation and range direction signal compression based on time-varying parameters is processed, and the range direction upsampling processing is performed. The method of range direction upsampling is frequency domain zero padding interpolation. Then, a column of echo signals after range direction pulse compression is read each time, and the signal corresponding to the slant range delay of the scene matching imaging grid point is projected onto the corresponding grid point. At the same time, the Doppler phase of the imaging grid point at this azimuth time is compensated, and the contribution of each pulse at this azimuth time is coherently accumulated, and finally the reconstruction result of each pixel point in the imaging grid can be obtained.

[0060]

[0061] Step five, phase preserving processing to obtain the final image.

[0062] In the back projection imaging processing, the Doppler phase of the target is removed for coherent superposition, and after completing the coherent superposition, the Doppler phase needs to be restored according to the slant range of the scene observation band center to ensure that the image phase corresponds to the corresponding physical position.

[0063]

[0064] To verify the advantages of the time-varying parameter based spaceborne scene matching SAR time domain imaging method, the parameters in Table 1 are used for simulation in strip mode, and the imaging ability is observed through the imaging result.

[0065] Step one, according to the parameters in table 1 and formula (5), the distance direction time-varying parameter phase compensation and quadratic phase compensation filter F is calculated RD The echo data s0 is subjected to distance direction Fourier transform to obtain the distance spectrum (5) of the signal, the distance direction time-varying parameter phase compensation and quadratic phase compensation filter F is calculated RD The echo signal is multiplied to obtain the distance spectrum of the filtered signal. Then, distance direction inverse Fourier transform is performed to obtain the distance direction compressed time domain signal s1. The imaging parameter list is shown in table 2.

[0066] Table 1 list of key parameters of SAR satellite

[0067]

[0068]

[0069] Table 2 list of imaging parameters

[0070]

[0071] Step two, the scene is divided based on radar beam pointing to match the imaging grid, as shown in Figure 3

[0072] Step three, the ground beam ellipse is accurately calculated to perform imaging grid point irradiation judgment, and the beam irradiation judgment ellipse is shown in Figure 4

[0073] Step four, backward projection imaging processing is performed.

[0074] Step five, phase preserving processing is performed to obtain the final image.

[0075] In this embodiment, point target and surface target simulation are used. Figure 5 Nine point target simulation results are given, and it can be found from the figure that the point target results of different scenes are well focused. Figure 6 Point target evaluation results are given, which show that the imaging method has good adaptability to time-varying parameters.

[0076] Figure 7 Surface target simulation results are given, and it can be found from Figure 7 that the imaging results are well focused and meet the index requirements.

[0077] In summary, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.​​

Claims

1. A space-borne SAR scene-matching mode time-domain imaging method, characterized in that, The method comprises the following steps: Step one, phase compensation is carried out on the target backscattering echo after distance Fourier transform, and signal distance spectrum is obtained; the signal distance spectrum is inversely Fourier transformed in the distance direction, and the signal compressed in the distance direction is obtained in the time domain; wherein the phase compensation method is: based on the second phase item of the target backscattering echo after distance Fourier transform, a distance time-varying parameter compensation and quadratic phase compensation filter F RD is generated; the target backscattering echo after distance Fourier transform is multiplied by F RD , and the signal distance spectrum is obtained; Step two, dividing the scene matching imaging grid based on the radar beam pointing; Step three, calculating the ground beam ellipse to determine the irradiation of the imaging grid points, specifically as follows: Selecting the corresponding azimuth time, moving the origin of the coordinate system to the position of the radar platform at the corresponding time; rotating the coordinate system around the Z axis, rotating the positive direction of the Y axis to the direction of the beam center pointing to the ground projection point; rotating the coordinate system around the X axis, pointing the positive direction of the Z axis to the ground beam center; determining the beam center rotation angle according to the attitude information of the radar satellite, rotating the Z axis by the corresponding angle; establishing an elliptical cone equation according to the two-dimensional beam width of the radar beam to determine the beam irradiation target of the imaging grid points, specifically as follows: If the Z value of the coordinate of the imaging grid point in the transformed coordinate system is greater than the X coordinate and the Y coordinate corresponding to the Z value on the elliptical cone, it is considered that the target is inside the elliptical cone and within the beam irradiation range, otherwise it is outside the beam irradiation range, obtaining the information of whether the imaging grid point is irradiated at each time, and obtaining the calculated ground beam ellipse template, the size of which is the maximum rectangular imaging grid covered by the beam; selecting a certain number of position beam centers to obtain the corresponding ground beam ellipse template; in the back projection process, different ground beam ellipse templates are selected according to the beam center at different positions; Step four, performing range upsampling processing on the signals after phase compensation and range compression processing; then, reading one column of echo signals that have been range pulse compressed each time, projecting the signals corresponding to the slant range delay of the scene matching imaging grid points onto the corresponding grid points, compensating the Doppler phase of the imaging grid point at the azimuth time, coherently accumulating the contribution of each pulse at the azimuth time, and finally obtaining the reconstruction result of each pixel point in the imaging grid; Step five, performing phase preserving processing on the reconstruction result obtained in step four to obtain the final image. In step five, the phase preserving processing is specifically: restoring the Doppler phase according to the slant range of the scene observation band center to ensure that the image phase corresponds to the corresponding physical position.

2. The method of claim 1, wherein, In step two, the beam pointing is obtained based on the actual radar satellite three-axis pointing in the satellite return auxiliary data, the wave foot is calculated combined with the orbit information to obtain the wave foot trend, and the scene matching imaging grid is realized according to the wave foot trend.

3. The method of claim 2, wherein, In step two, the way to realize the scene matching imaging grid according to the wave foot trend is to determine the scene observation band extension direction, and to establish the imaging grid along the scene observation band extension direction according to the resolution requirement.

4. The method of claim 1, 2 or 3, wherein, In step four, the method of range upsampling is frequency domain zero padding interpolation.

Images