CN-116203560-B - Space-borne large strabismus high-resolution sliding SAR frequency domain imaging method

CN116203560BCN 116203560 BCN116203560 BCN 116203560BCN-116203560-B

Abstract

The invention provides a satellite-borne large squint high-resolution sliding SAR frequency domain imaging method, which realizes high-precision imaging of the satellite-borne large squint high-resolution sliding SAR, solves the problem that the Doppler center of each target of an azimuth varies nonlinearly along with azimuth time through nonlinear declivity, solves the problem that azimuth space variables of range migration are inconsistent at different range gates through azimuth resampling, and overcomes the defects of the prior art.

Inventors

DING ZEGANG
LI HAN
Zheng Pengnan
ZHANG TIANYI
LI ZHE

Assignees

北京理工大学

Dates

Publication Date: 20260505
Application Date: 20230117

Claims (6)

1. The satellite-borne large strabismus high-resolution sliding SAR frequency domain imaging method is characterized by comprising the following steps of: step S1, establishing a satellite-borne large strabismus smooth SAR signal model; step S2, preprocessing, including distance walking correction and nonlinear azimuth declivity, to obtain a decoupled and unambiguous azimuth spectrum; S3, performing range migration azimuth space-variant correction, including azimuth resampling and polynomial compensation, wherein azimuth resampling removes the inconsistency of azimuth space variables of range migration at different range gates, and polynomial compensation removes consistent azimuth space variables of range migration; S4, performing range migration correction and range compression, adopting a range direction CS algorithm to realize range space variant correction of range migration, and completing range direction focusing; And S5, azimuth compression, namely removing azimuth null variables of the focusing parameters by adopting an azimuth NCS algorithm, and completing azimuth focusing to obtain a two-dimensional focusing image.
2. The method of space-borne large strabismus high-resolution sliding SAR frequency domain imaging according to claim 1, wherein said method of step S1 comprises: according to the geometric configuration of the spaceborne large squint smooth aggregation SAR, the range history of the target Expressed in taylor series expansion diagonal model: In the formula, In order to be the azimuth time, For the synthetic aperture center instant of the target, Is that The distance between the radar and the target at the moment, ~ The coefficients of each order of the pitch model are developed for the taylor series and it should be noted that, ~ With the slant distance of the target The synthetic aperture center moment of the target A change; based on the taylor series expansion model, the echo signal can be expressed as: In the formula, And The distance-wise and azimuth-wise signal envelopes, In order to be a distance-time, In order to achieve the light velocity, the light beam is, In order to achieve a frequency modulation rate, Is the wavelength.
3. The method of space-borne large strabismus high-resolution sliding SAR frequency domain imaging according to claim 2, wherein said method of step S2 comprises: first, the echo is subjected to a distance Fourier transform and the phase is compensated And (3) finishing distance walking correction: In the formula, For the radar carrier frequency, In order to be a distance frequency, In order to be the satellite velocity, Is the central oblique view angle; Then, compensate the phase Finishing non-linear azimuth declivity; In the formula, In the formula, 、 And Respectively, the Doppler centers of targets at different points in azimuth along with the synthetic aperture center moment of the targets First, second and third order rates of change.
4. The method of space-borne large strabismus high-resolution sliding SAR frequency domain imaging according to claim 3, wherein said method of step S3 comprises: after pretreatment, the pitch history of each target becomes: In the formula, Coefficients of each order of skew for different point targets in a scene ~ Is modeled by the space-variant of (3): In the formula, For the shortest skew of the scene center reference point, ~ For each order coefficient of the skew of the reference point, And Respectively is Along the distance and azimuth directions The derivative of the order, , ; First, distance compression is performed to compensate for phase And performing inverse distance Fourier transform: Wherein, the For frequency adjustment; then, azimuth resampling is carried out, the resampling coefficient changes along with the distance gate, and the relationship between azimuth time before and after resampling is as follows: In the formula, The azimuth time after the resampling is used, Can be expressed as: After interpolation, the distance Fourier transform is performed, and the phase is compensated Restoring the frequency modulation of the chirp signal; and finally, performing consistent range migration azimuth space-variant correction and compensating the phase: In the formula, 。
5. The method of space-borne large strabismus high-resolution sliding SAR frequency domain imaging according to claim 4, wherein said method of step S4 comprises: First, a signal expression of a two-dimensional frequency domain is obtained: In the formula, Is azimuth frequency; Will be described above in Taylor expansion, get: In the formula, ~ Taylor expansion coefficients for each order; Then, neglecting the null variability of the cubic term, compensating for phase : Wherein, the Representing the third-order Taylor expansion coefficient in A value of time; performing distance inverse Fourier transform to compensate phase Expressed as: In the formula, In the formula, Is that Derivative along the distance direction: Performing distance Fourier transform to compensate phase Then, performing inverse distance Fourier transform to complete distance migration correction and distance compression; Finally, compensating the residual phase : 。
6. The method of space-borne large strabismus high-resolution sliding SAR frequency domain imaging according to claim 5, wherein said method of step S5 comprises: first, the azimuth modulation is carried out in the phase Taylor expansion, get: In the formula, ~ For each order of the taylor expansion coefficient, Is that In azimuth direction The derivative of the order, , ; Then, compensating the phase in the azimuth frequency domain : In the formula, Wherein, the Is constant and is ; After the azimuth inverse Fourier transform, the phase is compensated : In the formula, Performing azimuth Fourier transform and compensating phase : In the formula, And finally, carrying out azimuth inverse Fourier transform to obtain a two-dimensional focusing SAR image.

Description

Space-borne large strabismus high-resolution sliding SAR frequency domain imaging method Technical Field The invention belongs to the technical field of synthetic aperture radars, and particularly relates to a satellite-borne large squint high-resolution sliding SAR frequency domain imaging method. Background The synthetic aperture radar (SYNTHETIC APERTURE RADAR, SAR) is a widely applied earth remote sensing technology, and has wide application prospects in the fields of disaster early warning, environment monitoring, military reconnaissance and the like. With the development of the spaceborne SAR, coverage performance and revisitation performance have become two important indexes of the spaceborne SAR. The coverage and revisiting performance of the spaceborne large squint SAR can be effectively improved through front-squint or back-squint irradiation of the wave beam. Meanwhile, a sliding SAR is a typical high resolution imaging mode of a satellite SAR. Therefore, on-board large strabismus high-resolution slide-focus SAR has become a research hotspot for current on-board SAR. However, with the improvement of resolution, the following two difficulties exist in on-board large strabismus high-resolution slide SAR imaging. First, the rotation of the sliding beam causes the Doppler center of the target with different azimuth points to change nonlinearly along with azimuth time, and the traditional algorithm ignores the nonlinear variation, so that spectrum aliasing is caused. Secondly, the range migration has two-dimensional space variation, and the azimuth space variation of the range migration is inconsistent at different range gates, so that the traditional algorithm ignores the inconsistency, and the range migration is failed to correct and defocuses the image. Disclosure of Invention In order to solve the problems, the invention provides a frequency domain imaging method of a satellite-borne large squint high-resolution sliding SAR, which can realize high-precision imaging of the satellite-borne large squint high-resolution sliding SAR. A satellite-borne large strabismus high-resolution sliding SAR frequency domain imaging method comprises the following steps: step S1, establishing a satellite-borne large strabismus smooth SAR signal model; step S2, preprocessing, including distance walking correction and nonlinear azimuth declivity, to obtain a decoupled and unambiguous azimuth spectrum; S3, performing range migration azimuth space-variant correction, including azimuth resampling and polynomial compensation, wherein azimuth resampling removes the inconsistency of azimuth space variables of range migration at different range gates, and polynomial compensation removes consistent azimuth space variables of range migration; S4, performing range migration correction and range compression, adopting a range direction CS algorithm to realize range space variant correction of range migration, and completing range direction focusing; And S5, azimuth compression, namely removing azimuth null variables of the focusing parameters by adopting an azimuth NCS algorithm, and completing azimuth focusing to obtain a two-dimensional focusing image. Preferably, the method of step S1 includes: Depending on the geometry of the on-board large squint SAR, the target's range history R (t a) can be represented by a Taylor series expansion range model: R(ta)＝r0+k1(ta-tp)+k2(ta-tp)2+k3(ta-tp)3+k4(ta-tp)4+k5(ta-tp)5 Where t a is azimuth time, t p is the synthetic aperture center time of the target, r 0 is the distance between the radar and the target at time t p, and k 1～k5 is each order coefficient of the taylor series expansion slope model. It should be noted that, k 1～k5 varies with the slant distance r 0 of the target and the synthetic aperture center time t p of the target; based on the taylor series expansion model, the echo signal can be expressed as: Where u r (·) and u a (·) are the distance and azimuth signal envelopes, respectively, t r is distance time, c is speed of light, K r is frequency modulation, and λ is wavelength. Preferably, the method of step S2 includes: firstly, performing distance Fourier transform on the echo, compensating phase H 1, and completing distance walk correction: Wherein f c is radar carrier frequency, f r is distance frequency, v s is satellite speed, Is the central oblique view angle; then, compensating the phase H 2 to finish non-linear azimuth declivity; In the formula, Wherein, f dc1、fdc2 and f dc3 are the first, second and third order change rates of the Doppler center of the target at different points in azimuth along with the time t p of the synthetic aperture center of the target respectively. Preferably, the method of step S3 includes: after pretreatment, the pitch history of each target becomes: R(ta)＝R0+K2(ta-tp)2+K3(ta-tp)3+K4(ta-tp)4+K5(ta-tp)5 In the formula, Modeling the space-variant of each-order coefficient K 2～K5 of the skew of targets at different points in a scene: Wherein R