A three-dimensional deformation inversion method for navigation satellite interferometric SAR

Abstract

The present invention provides a three-dimensional deformation inversion method for navigation satellite interferometric SAR, which solves the problem of experimental acquisition time and selection of launching satellites. It includes: calculating the average resolution unit area of ​​the target scene within a re-orbit cycle based on prior trajectory information; calculating the satellite elevation angle at each moment, and filtering out satellites with elevation angles less than a set value; obtaining a combination of acquisition time and available satellites that meet the conditions according to the average resolution unit area of ​​the scene and the result of screening based on the satellite elevation angle; obtaining images of each satellite at this time according to the obtained acquisition time and available satellite combination that meet the conditions, and obtaining independent points in the image based on the shape of the resolution unit; for the independent points, obtaining all valid independent point sets in the image by calculating the signal-to-noise ratio and the sidelobe interference of adjacent points, and selecting the satellite combination and acquisition time with the largest number of valid independent points; performing extreme point selection processing on the SAR image set in turn, and obtaining a set of valid independent points.

Description

Technical Field

[0001] The invention belongs to the technical field of bistatic synthetic aperture radar, and in particular relates to a three-dimensional deformation inversion method of a navigation satellite interferometric SAR. Background Art

[0002] The interferometric bistatic synthetic aperture radar based on the navigation satellite system (GNSS-InBSAR) is a bistatic SAR system that uses navigation satellites as external radiation sources and arranges receivers on the ground or near the ground to receive echoes from the target scene. It has outstanding advantages such as short re-orbit time, wide coverage, continuous monitoring in time and space, and low cost.

[0003] However, due to the large number of satellites in orbit in the navigation satellite system, there are many choices and combinations of satellites to launch in the three-dimensional deformation detection experiment, and the best combination needs to be selected to achieve the best deformation detection results; in addition, the system's joint configuration, resolution, elevation angle and imaging results will affect the final three-dimensional deformation detection accuracy. Therefore, the experimental design problem for the navigation satellite system needs to be solved. Summary of the invention

[0004] In view of this, the present invention provides a three-dimensional deformation inversion method for navigation satellite interferometric SAR, which solves the problems of experimental acquisition time and selection of launching satellites, and provides a prerequisite for obtaining high-precision deformation variables.

[0005] The present invention is achieved through the following technical solutions.

[0006] A three-dimensional deformation inversion method for navigation satellite interferometric SAR, comprising:

[0007] Based on the prior trajectory information, the average resolution unit area of ​​the target scene within a re-orbit cycle is calculated; at the same time, the satellite elevation angle at each moment is calculated, and the satellites with elevation angles less than the set value at each moment are filtered out;

[0008] According to the average resolution unit area of ​​the scene and the results of screening based on the satellite elevation angle, the optimal PDOP output under different satellite combinations is calculated to obtain the acquisition time and available satellite combination that meets the conditions;

[0009] According to the obtained acquisition time that meets the conditions and the combination of available satellites, the image of each satellite at this time is obtained, and the independent point in the image is obtained based on the shape of the resolution unit;

[0010] For the independent points, all valid independent point sets in the image are obtained by calculating the signal-to-noise ratio and the sidelobe interference of adjacent points, and the satellite combination and acquisition time with the largest number of valid independent points are selected;

[0011] The extreme point selection process is performed on the SAR image set in turn, and a set of effective independent points is obtained.

[0012] Beneficial effects of the present invention:

[0013] Based on prior information, the present invention selects the acquisition time that meets the requirements of imaging and deformation inversion from three aspects: theoretical resolution, elevation angle and position precision factor (PDOP). Then, based on the imaging results at different times, the optimal acquisition time is obtained by calculating the number of effective observation points, thereby realizing the design of three-dimensional deformation inversion experiments, which plays an important role in the practical application of GNSS-InBSAR systems. DETAILED DESCRIPTION

[0014] The present invention is described in further detail below.

[0015] The three-dimensional deformation inversion method of the navigation satellite interferometer SAR in this specific implementation is specifically set as follows: the navigation satellite interferometer SAR system adopts the northeast sky coordinate system, the transmitter is the Beidou IGSO and MEO satellite, and its pitch angle It is defined as the angle with the horizontal plane, which is positive when it is upward. The azimuth angle θ is defined as the angle with the due east direction, which is positive when it is counterclockwise. The receiver is stationary and placed at the origin of the system. The receiver has two antennas, which receive echo signals and direct wave signals respectively.

[0016] Step 1: Based on the prior trajectory information, the average resolution unit area of ​​the target scene within a heavy track period is calculated;

[0017] Compared with the traditional chirp signal, the transmission signal of the navigation star system is a CA code. After the pulse compression in the range direction, it is a triangle wave, while the pulse compression result in the azimuth direction is still a sinc signal. In addition, considering the spatial variation of the resolution in the scene, under the condition that the receiver is stationary, the projection of the system's range and azimuth resolution on the ground is:

[0018]

[0019] Where B is the CA code signal bandwidth, T int is the synthetic aperture time, β is the bistatic angle, Θ is the component in the direction of the β bisector, ω TA is the angular velocity of the transmitter relative to the point target A, TA is the unit vector of the transmitter in the effective direction of motion, and c is the speed of light. Ω r ,Ω a for:

[0020] Ω r =Ξ×Z

[0021] Ω a =θ×Z

[0022] Where × represents the cross product, Ξ is the equivalent motion direction, and Z is the unit vector pointing upwards;

[0023] Therefore, the resolution unit area corresponding to a single target is approximately:

[0024] S reso =ρ a ρ r cosφ

[0025] Where φ is Ω r ,Ω a The angle between.

[0026] According to the prior trajectory information, the area of ​​the resolution unit corresponding to each moment is calculated. In the PS-InSAR processing, in order to increase the number of PS points as much as possible and reduce the mutual phase influence between PS points, the area of ​​the resolution unit should be as small as possible. When the transmission waveform and synthetic aperture time are constant, the area of ​​the scene itself is very small compared to the distance from the satellite to the scene, and it can be approximately considered that Φ TA and Γ TA Therefore, the resolution of the scene is mainly related to the angle between the target and the receiver.

[0027] This example uses the trajectory information of Beidou2_IGSO3 at 2:00 p.m. on May 3, 2021 as an example to simulate the resolution of a 1200*1200 simulation area. The receiver is set at the origin. The results show that the resolution change is only related to the angle. Therefore, when calculating the scene resolution, the average resolution at different angles within the antenna main lobe width should be calculated based on the illumination range of the receiving antenna to achieve the best resolution for the entire scene.

[0028] Step 2: Calculate the satellite elevation angle at each moment, and filter out the satellites whose elevation angles are less than the set value at each moment;

[0029] In this embodiment, the calculation of the satellite elevation angle at each moment is specifically as follows:

[0030] For position P s =[P sx ,P sy ,P sz ], the elevation angle is:

[0031]

[0032] In specific implementation, if the elevation angle is too low, the distance that the signal passes through the atmosphere will increase, thereby affecting the signal-to-noise ratio of the echo signal and increasing the phase noise of the received signal. Therefore, in this embodiment, it is set that satellite signals with elevation angles below 15° cannot be used. The satellite elevation angle at each moment is obtained through prior information, and then satellites with lower elevation angles are screened out.

[0033] Step 3: Calculate the optimal PDOP output under different satellite combinations according to the average resolution unit area of ​​the scene and the results after screening based on the satellite elevation angle, and obtain the acquisition time and available satellite combination that meets the conditions;

[0034] The PDOP (Position Dilution of Precision) is commonly used in the technical field to calculate the positioning accuracy of a navigation star system. The present invention expands the definition of PDOP and combines the correlation coefficients at different angles to calculate the theoretical accuracy of three-dimensional inversion.

[0035] In this embodiment, the PS-InSAR principle is used to realize deformation detection, and the error accuracy under multi-satellite observation is expressed as:

[0036]

[0037] Among them, σ w is the standard deviation of deformation detection, H M×3 is the deformation measurement matrix. Then, PDOP is defined as:

[0038]

[0039] The PDOP in each direction is defined as:

[0040]

[0041] In this embodiment, for Beidou IGSO satellites, the spatial coherence coefficient varies greatly within a re-orbit period, that is, within 1 day, so it is necessary to select the satellites used for observation, and select the optimal result by calculating the PDOP values ​​of different combinations. The final design results are shown in Table 1 below.

[0042] Table 1

[0043]

[0044] Step 4: According to the obtained acquisition time that meets the conditions and the combination of available satellites, the image of each satellite at this time is obtained, and the independent point in the image is obtained based on the shape of the resolution unit;

[0045] The independent points in this embodiment refer to target points with complete resolution units in the image, which are obtained in the following way:

[0046] Through steps 1 to 3, the collection time candidate set {t 1 ...t T There are T acquisition moments in total, and the corresponding N satellites used for deformation inversion {S 1 ...SN} and the SAR images corresponding to each satellite {I 1 ...I N}; then from I 1 First, according to the configuration and parameters of the navigation satellite interferometric SAR system, the theoretical PSF shape of each point is obtained. Taking -3dB as the threshold, for any target point A, the theoretical PSF is expressed as:

[0047] S(A) PSF ={B|χ(A,B)>-3dB}

[0048] Among them, χ(A,B) represents the fuzzy function between the target point A and its neighboring target vector B.

[0049] When the target point A is S(A) PSF If A is the maximum point in I, then A is considered to be an independent point. 1 , get all the extreme point sets M 1 %.

[0050] Step 5: For the independent points, all valid independent point sets in the image are obtained by calculating the signal-to-noise ratio and the sidelobe interference of adjacent points, and the satellite combination and acquisition time with the largest number of valid independent points are selected;

[0051] In this embodiment, the effective independent point set is obtained in the following manner:

[0052] 5.1 Using the set signal-to-noise ratio as the threshold, the extreme point set M 1 % of the points are screened to obtain an independent point set:

[0053]

[0054] Among them, SNR thre is the threshold value of signal-to-noise ratio;

[0055] 5.2 Calculate the influence of the target point A on Q according to the target point strength;

[0056] In specific implementation, step 5.1 obtains an independent point set in the scene that meets the signal-to-noise ratio requirement, but for the M 1 The point Q in % may be affected by the surrounding points, interfering with the accuracy of the phase. Therefore, this embodiment calculates the influence of the target point A on Q according to the target point strength:

[0057]

[0058] Among them, σ A is the intensity of the target point, that is, the amplitude in the SAR image.

[0059] 5.3 Calculate the total interference amplitude according to the influence amplitude of the target point A on Q;

[0060] Assuming that there are L interference points around point Q, the total interference amplitude of the L interference points on Q is:

[0061] Int(Q) = Amp(Q; A 1 )+L+Amp(Q;A L )

[0062] 5.4 According to the amplitude ratio threshold value Amp thre The total interference amplitude is screened to obtain a valid independent point set M 1 ;

[0063] For the target point Q, its amplitude should be large enough to ensure that it will not be disturbed, that is:

[0064]

[0065] Then the effective independent point set M 1 for:

[0066]

[0067] Step 6: Sequentially analyze the SAR image set {I 1 ...I N} to select extreme points and obtain a valid independent point set {M 1 ...M N}, then at the acquisition time t 1 , the number of valid independent points is

[0068] Traverse the collection time set {t 1 ...t T}, the number of effective independent points is {M t1 ...M tN}. In summary, the final sampling time is According to the design results of the collection time in step 3, the final effective independent point selection results at each time point can be obtained as shown in Table 2.

[0069] Table 2

[0070]

[0071] According to Table 2, the average number of valid points obtained by collecting data at 2:00 is the largest, so the final experimental time is 2:00, and the satellites Beidou2_IGSO1, Beidou2_IGSO4 and Beidou2_IGSO6 are selected.

[0072] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A three-dimensional deformation inversion method for navigation satellite interferometric SAR, characterized in that: include: Based on the prior trajectory information, the average resolution unit area of ​​the target scene within a heavy track period is calculated; At the same time, the satellite elevation angle at each moment is calculated, and the satellites whose elevation angle at each moment is less than the set value are filtered out; According to the average resolution unit area of ​​the scene and the results of screening based on the satellite elevation angle, the optimal PDOP output under different satellite combinations is calculated to obtain the acquisition time and available satellite combination that meets the conditions; According to the obtained acquisition time that meets the conditions and the combination of available satellites, the image of each satellite at this time is obtained, and the independent point in the image is obtained based on the shape of the resolution unit; the independent point is obtained in the following way: By collecting the time candidate set {t1...t T There are T acquisition moments in total, and the corresponding N satellites used for deformation inversion {S1...S N } and the SAR images corresponding to each satellite {I1...I N }; Starting from I1, according to the configuration and parameters of the navigation satellite interferometric SAR system, the theoretical PSF shape of each point is obtained. Taking -3dB as the threshold, for any target point A, the theoretical PSF is expressed as: S(A) PSF ={B|χ(A,B)>-3dB} Among them, χ(A,B) represents the fuzzy function between the target point A and its neighboring target vector B; When the target point A is S(A) PSF When it is a maximum point within , A is considered to be an independent point; For the independent points, all valid independent point sets in the image are obtained by calculating the signal-to-noise ratio and the sidelobe interference of adjacent points, and the satellite combination and acquisition time with the largest number of valid independent points are selected; The extreme point selection process is performed on the SAR image set in turn, and a set of effective independent points is obtained.

2. The three-dimensional deformation inversion method of navigation satellite interferometric SAR according to claim 1, characterized in that: The resolution unit area is calculated in the following way: S reso =ρ a r r cosφ Where φ is Ω r ,Ω a The angle between × represents the cross product, Θ is the component in the direction of the β bisector, β is the bistatic angle, Ξ is the equivalent motion direction, Z is the unit vector pointing upward, and ρ α Represents the projection of the azimuth resolution on the ground, ρ γ Represents the projection of the system's range resolution onto the ground.

3. The three-dimensional deformation inversion method of navigation satellite interferometric SAR according to claim 1 or 2, characterized in that: The calculation of the satellite elevation angle at each moment is specifically as follows: For position P s =[P sx ,P sy ,P sz ], the satellite elevation angle is:

4. The three-dimensional deformation inversion method of navigation satellite interferometric SAR according to claim 3, characterized in that: The set value of the satellite elevation angle is 15°.

5. The three-dimensional deformation inversion method of navigation satellite interferometric SAR according to claim 1 or 2, characterized in that: After obtaining the independent points, further traverse I1 to obtain all extreme point sets.

6. The three-dimensional deformation inversion method of navigation satellite interferometric SAR according to claim 5, characterized in that: The effective independent point set is obtained in the following way: 5.1 Using the set signal-to-noise ratio as a threshold, the points of the extreme point set are screened to obtain an independent point set; 5.2 Calculate the influence of the target point A on the target point Q according to the target point strength; 5.3 Calculate the total interference amplitude according to the influence amplitude of the target point A on Q; 5.4 According to the amplitude ratio threshold value Amp thre The total interference amplitude is screened to obtain a valid independent point set.

Images