Imaging method of missile-borne SAR to correct large scene range migration based on CS algorithm
Through the bomb-load SAR correction method based on compression perception algorithm, the problem that traditional algorithms cannot effectively correct the distance migration of long-distance targets in large-scene imaging is solved, and more accurate distance migration correction and orientation focusing effects are achieved.
Patent Information
- Application Number
- CN202210339383.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-01
- Publication Date
- 2025-02-14
- Estimated Expiration
- 2042-04-01
AI Technical Summary
The existing bomb-based SAR distance Doppler algorithm cannot effectively correct the distance migration of long-distance targets in large-scene imaging, resulting in large imaging errors.
Using the bomb-load SAR correction method based on compression perception (CS) algorithm, the two-dimensional spectrum expression of echo data is derived through series inversion method, and Fourier transforms of azimuth and distance directions are performed, combining Taylor expansion and modulation The scale function performs distance migration correction and orientation focus.
Under the large-scene model, the distance movement of long-distance targets can be effectively corrected, the orientation focus effect can be improved, the accurate imaging of targets of each distance unit can be ensured, and the imaging error can be reduced.
Smart Images
Figure CN114660604B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a CS method for correcting large scene range migration of missile-borne SAR. Background Art
[0002] In the process of missile-borne SAR imaging of scenes, since synthetic aperture radar transmits signals with a wide frequency band in a large array, there will be inevitable coupling between the distance dimension and the azimuth dimension of the scene echo signal, namely, the range migration in the imaging system, which makes the system response present as a two-dimensional curve in the fast time domain and the slow time domain. At the same time, compared with traditional airborne SAR, the missile-borne platform has more complex motion characteristics: fast speed, non-uniform speed change, and complex changes in the slant distance between the radar and the target.
[0003] Although the traditional airborne SAR imaging algorithm takes into account the spatial variability of the target distance in the radar platform detection scene, for the missile-borne SAR imaging model, the extremely fast flight speed and complex motion characteristics of the missile platform cause the target range migration to be very different from the conventional SAR model, and the traditional SAR imaging algorithm cannot be used directly. As for the existing missile-borne SAR range Doppler (RD) algorithm, it also uses the series inversion method to derive the two-dimensional spectrum expression of the target echo data, but when processing the two-dimensional spectrum, the target position variable is usually replaced by the coordinates of the scene center point. This approximation can perform range migration correction and azimuth focusing relatively well for targets near the scene center line, but there are large errors for targets at other locations far from the scene center under the large scene model. Therefore, when the radar imaging mapping swath width is extended to a large scene, the change in the position coordinate distance difference will directly lead to over-correction or under-correction of range migration and azimuth defocusing of targets far from the scene center line. Summary of the invention
[0004] The purpose of the present invention is to solve the problem that the existing missile-borne SAR range Doppler algorithm is not suitable for wide mapping band and large scene imaging, resulting in large errors in target imaging, and propose an imaging method for missile-borne SAR to correct large scene range migration based on CS algorithm.
[0005] The specific process of the missile-borne SAR imaging method for correcting large scene range migration based on the CS algorithm is as follows:
[0006] Step 1: The radar samples the echo of the scene target to obtain the original data, and uses the series inversion method to derive the two-dimensional spectrum expression of the echo data;
[0007] Step 2: Raw data Perform azimuth FFT to transform the original data from the time domain Transform to azimuth Doppler domain get
[0008] The FFT is a fast Fourier transform; is the distance to time domain time; t m is the time in the azimuth domain; R0 is the instantaneous distance from the radar to the target; is the data after Fourier transformation of the original data; ΔR is the distance position offset of the target relative to the center of the scene; f a is the azimuth frequency domain;
[0009] Step 3: Substitute the total phase term Φ(f r ,f a ) The first-order coefficient in the distance frequency domain Perform Taylor expansion at ΔR=0 to construct the line frequency modulation variable scaling function H1;
[0010] Will Multiplying by H1 eliminates the spatial variation characteristics of the distance migration of all distance unit targets in the scene, and adjusts them all to be consistent with the distance migration of the center line of the scene, and obtains
[0011] Step 4: Compensate the data consistent with the one completed in step 3 Perform distance FFT transformation to two-dimensional frequency domain f r -f a , we get S3(f r ,f a ,R0);
[0012] S3(f r ,f a ,R0) is Data after distance Fourier transformation; f r is the distance frequency domain;
[0013] Step 5: Transform step 4 into the two-dimensional frequency domain f r -f a The data after S3(f r ,f a , R0) is multiplied by H2 to complete the distance migration correction and distance compression processing in the frequency domain, and the data after the distance migration correction and distance compression processing in the frequency domain are subjected to the distance IFFT to obtain
[0014] The H2 is a two-dimensional frequency domain system function; IFFT is an inverse Fourier transform; S3(f r ,f a , R0) after completing the range motion correction and pulse compression, the data is subjected to the range inverse Fourier transform;
[0015] Step 6: Multiply by H3, The processed data obtained by multiplying by H3 is subjected to azimuth IFFT to obtain the SAR image;
[0016] The H3 is the azimuth pulse pressure and phase compensation function; SAR is synthetic aperture radar.
[0017] The beneficial effects of the present invention are:
[0018] Under the large scene model, the distance difference change of the target position coordinates is taken into account to ensure that the range migration of the target in each range unit can be well corrected and the residual phase can be compensated to improve the azimuth focusing effect. The present invention proposes an imaging method for correcting large scene range migration of missile-borne SAR based on the CS algorithm.
[0019] Under the same large-scene radar parameter model, such as the parameter model given by the present invention, the traditional missile-borne SAR range Doppler algorithm has poor range migration correction effect on long-distance targets, and the azimuth compression and broadening are serious, making it difficult to focus on point pulses; while the missile-borne SAR range migration correction method based on CS can perform good correction of the range migration of long-distance targets under the large-scene model, ensure the focusing effect in azimuth, and accurately image the targets in each distance unit in the scene. The problem that the existing missile-borne SAR range Doppler algorithm is not suitable for imaging of large scenes with wide surveying strips is solved, resulting in large errors in target imaging. In summary, the missile-borne SAR range migration correction imaging method based on the CS algorithm has better range migration correction and azimuth focusing effects in large scenes than the traditional missile-borne SAR range Doppler algorithm. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a large scene front and side view geometric model diagram used in the present invention. A and B are the missile platform routes. m = 0, the point projected onto the ground by the radar platform is the origin O, the radar route direction is the OX direction, the vertical radar route direction is the OY direction, the OZ direction is perpendicular to the OX direction and the OY direction, P1 is the short-range point in the imaging scene, and P2 is the long-range point in the imaging scene; O′ is the intersection of the missile platform route on the OZ axis;
[0021] Figure 2 This is a flowchart of missile-borne SAR imaging based on the CS algorithm of the present invention;
[0022] Figure 3a It is the uncorrected distance compression graph of the distance migration of the traditional RD algorithm;
[0023] Figure 3b It is the distance compression graph after distance migration correction of the traditional RD algorithm;
[0024] Figure 4aThis is the uncorrected range compression graph of the long-distance target range migration of the traditional RD algorithm;
[0025] Figure 4b This is the distance compression diagram after the traditional RD algorithm corrects the distance migration of the long-distance target;
[0026] Figure 5 It is the azimuth compression graph of the traditional RD algorithm;
[0027] Figure 6 This is the three-dimensional imaging diagram of the traditional RD algorithm;
[0028] Figure 7a It is the uncorrected distance compression graph of the CS algorithm distance migration of the present invention;
[0029] Figure 7b It is the distance compression map after the distance migration correction of the CS algorithm of the present invention;
[0030] Figure 8a It is the uncorrected distance compression diagram of the distance migration of the long-distance target by the CS algorithm of the present invention;
[0031] Figure 8b It is the distance compression diagram after the distance migration correction of the long-distance target by the CS algorithm of the present invention;
[0032] Fig. 9 It is the CS algorithm azimuth compression diagram of the present invention;
[0033] Fig.10 Three-dimensional imaging diagram of the CS algorithm of the present invention. DETAILED DESCRIPTION
[0034] Specific implementation method 1: Combination Figure 1 The present embodiment is described. The specific process of the imaging method of the missile-borne SAR based on the CS algorithm in the present embodiment for correcting the large scene range migration is as follows:
[0035] Step 1: The radar samples the echo of the scene target to obtain the original data, and uses the series inversion method to derive the two-dimensional spectrum expression of the echo data; the MATLAB tool is used to sample the echo data to obtain the original echo; the model is a missile-borne SAR, and the radar is mounted on the missile;
[0036] Step 2: Raw data Perform azimuth FFT to transform the original data from the time domain Transform to azimuth Doppler domain get
[0037] The FFT is a fast Fourier transform; is the distance to time domain time; t m is the time in the azimuth domain; R0 is the instantaneous distance from the radar to the target; is the data after Fourier transformation of the original data; ΔR is the distance position offset of the target relative to the center of the scene; f a is the azimuth frequency domain;
[0038] Step 3: Substitute the total phase term Φ(f r ,f a ) The first-order coefficient in the distance frequency domain Perform Taylor expansion at ΔR=0 to construct the line frequency modulation variable scaling function H1;
[0039] Will Multiplying by H1 eliminates the spatial variation characteristics of the distance migration of all distance unit targets in the scene, and adjusts them all to be consistent with the distance migration of the center line of the scene, and obtains
[0040] Step 4: Compensate the data consistent with the one completed in step 3 Perform distance FFT transformation to two-dimensional frequency domain f r -f a , we get S3(f r ,f a ,R0);
[0041] S3(f r ,f a ,R0) is Data after distance Fourier transformation; f r is the distance frequency domain;
[0042] Step 5: Transform step 4 into the two-dimensional frequency domain f r -f a The data after S3(f r ,f a , R0) is multiplied by H2 to complete the distance migration correction and distance compression processing in the frequency domain, and the data after the distance migration correction and distance compression processing in the frequency domain are subjected to the distance IFFT to obtain
[0043] The H2 is a two-dimensional frequency domain system function; IFFT is an inverse Fourier transform; S3(f r ,f a , R0) after completing the range motion correction and pulse compression, the data is subjected to the range inverse Fourier transform;
[0044] Step 6: Multiply by H3, The processed data obtained by multiplying by H3 is subjected to azimuth IFFT to obtain the SAR image, and the entire process of this imaging processing is completed;
[0045] The H3 is the azimuth pulse pressure and phase compensation function; SAR is synthetic aperture radar.
[0046] Specific implementation method 2: This implementation method is different from specific implementation method 1 in that in step 1, the radar samples the scene target echo to obtain raw data, and uses the series inversion method to derive the two-dimensional spectrum expression of the echo data; the MATLAB tool is used to sample the echo data to obtain the original echo; the model is a missile-borne SAR, and the radar is mounted on the missile; the specific process is:
[0047] With t m = 0, the point projected onto the ground by the radar platform is the origin O, the radar flight path direction is the OX direction, the vertical radar flight path direction is the OY direction, and the OZ direction is perpendicular to the OX direction and the OY direction;
[0048] Suppose at a certain time t m , the slant range between the radar and the point target P is
[0049]
[0050] Where H is the height of the radar platform; V x0 and a x are the velocity and acceleration components of the radar platform (i.e., the missile) in the X-axis direction; V z0 and a z are the velocity and acceleration components of the radar platform (i.e., the missile) in the Z-axis direction; p is the distance from the target point P to the X axis; ΔR is the distance position offset of the target relative to the center of the scene; x p is the distance from the target point P to the Y axis;
[0051] Apply the slope distance formula at t m = 0 and perform third-order Taylor expansion, we can get
[0052]
[0053] Among them, R cen t m = 0 when the radar reaches the target P, b1b2 b3 are the coefficients of the first, second and third terms respectively;
[0054] The radar transmits a linear frequency modulation (LFM) signal, and the radar samples the echo of the scene target to obtain the raw data Using the series inversion method, the two-dimensional spectrum expression of the echo signal can be obtained as follows:
[0055]
[0056]
[0057] Among them, a r (·), a a (·) are the window functions of the echo signal in range and azimuth respectively; λ is the signal wavelength; f r is the distance frequency domain; f a is the azimuth frequency domain (i.e. Doppler domain); Φ(f r ,f a ) is the total phase term of the two-dimensional spectrum, which can be decomposed into the form of formula (4); are the constant term, first-order term and second-order term coefficients in the distance frequency domain in the total phase term of the two-dimensional spectrum; j is the imaginary unit, j 2 =-1.
[0058] The other steps and parameters are the same as those in the first embodiment.
[0059] Specific implementation method three: This implementation method is different from specific implementation method one or two in that in step three, the total phase term Φ(f r ,f a ) The first-order coefficient in the distance frequency domain Perform Taylor expansion at ΔR=0 to construct the line frequency modulation variable scaling function H1;
[0060] Will Multiplying by H1 eliminates the spatial variation characteristics of the distance migration of all distance unit targets in the scene, and adjusts them all to be consistent with the distance migration of the center line of the scene, and obtains
[0061] The specific process is:
[0062]
[0063]
[0064]
[0065] Where H1 is the line frequency modulation function; c is the speed of light, which is 3×10 8 m / s; A(f a ) is the coefficient of the first-order term The constant term of B(f a ) is the CS factor; for The data after distance migration compensation is consistent; f c is the carrier frequency of the LFM signal.
[0066] The other steps and parameters are the same as those in the first or second embodiment.
[0067] Specific implementation method 4: This implementation method is different from any one of the specific implementation methods 1 to 3 in that the two-dimensional frequency domain system function H2 in step 5 is:
[0068]
[0069] Wherein, γ is the frequency modulation slope of the LFM signal.
[0070] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.
[0071] Specific implementation method 5: This implementation method is different from the specific implementation methods 1 to 4 in that the azimuth pulse pressure and phase compensation function H3 in step 6 is:
[0072]
[0073] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.
[0074] The following examples are used to verify the beneficial effects of the present invention:
[0075] Embodiment 1:
[0076] The imaging method for correcting range migration of missile-borne SAR based on the CS algorithm in this embodiment is specifically prepared according to the following steps:
[0077] A parametric model is now provided:
[0078] Table 1 Basic parameter values of missile-borne SAR imaging method based on CS algorithm
[0079]
[0080]
[0081] The simulation results of point targets are shown in Figure 3a , 3b ,4a,4b,5,6,7a,7b,8a,8b,9,10; Figure 3a , 3b This is a comparison chart of distance migration correction of the traditional RD algorithm; Figure 4a , 4b This is a comparison chart of the range migration correction of long-distance targets using the traditional RD algorithm; Figure 5 It is the azimuth compression graph of the traditional RD algorithm; Figure 6 This is the three-dimensional imaging diagram of the traditional RD algorithm; Figure 7a , 7b This is a comparison diagram of the range migration correction of the CS algorithm of the present invention; Figure 8a , 8b This is a comparison chart of the long-distance target range migration correction using the CS algorithm of the present invention; Fig. 9It is the CS algorithm azimuth compression diagram of the present invention; Fig.10 This is the 3D imaging diagram of the CS algorithm of the present invention. Figure 3a , 3b and Figure 7a , 7b , Figure 4a , 4b and Figure 8a , 8b The comparison results show that the distance migration correction effect of the method of the present invention is better than that of the traditional method. Figure 5 and Fig. 9 The comparison results show that the azimuth focusing effect of the method of the present invention is better than that of the traditional method. Fig.10 and Figure 6 The comparison results show that the scene imaging effect of the method of the present invention is better than that of the traditional method.
[0082] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.
Claims
1. An imaging method for missile-borne SAR correction of large scene range migration based on CS algorithm, characterized by: The specific process of the method is: Step 1: The radar samples the echo of the scene target to obtain the original data, and uses the series inversion method to derive the two-dimensional spectrum expression of the echo data; Step 2: Raw data Perform azimuth FFT to transform the original data from the time domain Transform to azimuth Doppler domain get The FFT is a fast Fourier transform; is the distance to time domain time; t m is the time in the azimuth domain; R0 is the instantaneous distance from the radar to the target; is the data after Fourier transformation of the original data; ΔR is the distance position offset of the target relative to the center of the scene; f a is the azimuth frequency domain; Step 3: Substitute the total phase term Φ(f r ,f a ) The first-order coefficient in the distance frequency domain Perform Taylor expansion at ΔR=0 to construct the line frequency modulation variable scaling function H1; Will Multiplying by H1 eliminates the spatial variation characteristics of the distance migration of all distance unit targets in the scene, and adjusts them all to be consistent with the distance migration of the center line of the scene, and obtains Step 4: Compensate the data consistent with the one completed in step 3 Perform distance FFT transformation to two-dimensional frequency domain f r -f a , we get S3(f r ,f a ,R0); S3(f r ,f a ,R0) is Data after distance Fourier transformation; f r is the distance frequency domain; Step 5: Transform step 4 into the two-dimensional frequency domain f r -f a The data after S3(f r ,f a , R0) is multiplied by H2 to complete the distance migration correction and distance compression processing in the frequency domain, and the data after the distance migration correction and distance compression processing in the frequency domain are subjected to the distance IFFT to obtain The H2 is a two-dimensional frequency domain system function; IFFT is an inverse Fourier transform; S3(f r ,f a , R0) after completing the range motion correction and pulse compression, the data is subjected to the range inverse Fourier transform; Step 6: Multiply by H3, The processed data obtained by multiplying by H3 is subjected to azimuth IFFT to obtain the SAR image; H3 is the azimuth pulse pressure and phase compensation function; SAR is synthetic aperture radar; In step 1, the radar samples the scene target echo, obtains the original data, and uses the series inversion method to derive the two-dimensional spectrum expression of the echo data; the specific process is: With t m = 0, the point projected onto the ground by the radar platform is the origin O, the radar flight direction is the OX direction, the vertical radar flight direction is the OY direction, and the OZ direction is perpendicular to the OX direction and the OY direction; Suppose at a certain time t m , the slant range between the radar and the point target P is Where H is the height of the radar platform; V x0 and a x are the velocity and acceleration components of the radar platform in the X-axis direction respectively; V z0 and a z are the velocity and acceleration components of the radar platform in the Z-axis direction respectively; p is the distance from the target point P to the X axis; ΔR is the distance position offset of the target relative to the center of the scene; x p is the distance from the target point P to the Y axis; Apply the slope distance formula at t m = 0 and perform third-order Taylor expansion, we can get Among them, R cen t m = 0 when the radar reaches the target P, b1b2 b3 are the coefficients of the first, second and third terms respectively; The radar transmits a linear frequency modulation signal, and the radar samples the echo of the scene target to obtain the original data Using the series inversion method, the two-dimensional spectrum expression of the echo signal can be obtained as follows: Among them, a r (·), a a (·) are the window functions of the echo signal in range and azimuth respectively; λ is the signal wavelength; f r is the distance frequency domain; f a is the azimuth frequency domain; Φ(f r ,f a ) is the total phase term of the two-dimensional spectrum; are the constant term, first-order term and second-order term coefficients in the distance frequency domain in the total phase term of the two-dimensional spectrum; j is the imaginary unit, j 2 = -1; In step 3, the total phase term Φ(f r ,f a ) The first-order coefficient in the distance frequency domain Perform Taylor expansion at ΔR=0 to construct the line frequency modulation variable scaling function H1; Will Multiplying by H1 eliminates the spatial variation characteristics of the distance migration of all distance unit targets in the scene, and adjusts them all to be consistent with the distance migration of the center line of the scene, and obtains The specific process is: Where H1 is the line frequency modulation function; c is the speed of light, which is 3×10 8 m / s; A(f a ) is the coefficient of the first-order term The constant term of B(f a ) is the CS factor; for The data after distance migration compensation is consistent; f c is the carrier frequency of the LFM signal; The two-dimensional frequency domain system function H2 in step 5 is: Where, γ is the frequency modulation slope of the LFM signal; The azimuth pulse pressure and phase compensation function H3 in step 6 is: