A high-precision imaging scheme for rotating scanning ground-based SAR based on fast BP algorithm

Abstract

The present invention relates to the combination of a rotating scanning ground-based SAR and a fast BP algorithm, and discloses a high-precision imaging scheme for a rotating scanning ground-based SAR based on a fast BP algorithm. In the scheme, the azimuth phase history function and the number of pulse cycles required for the movement across resolution units are first calculated based on the gridded elevation auxiliary data and stored in a data file in the form of a three-dimensional array; secondly, the echo signal sampling data of the monitoring area is recorded by arc rotating scanning, and for each modulation period, the data is subjected to spectrum analysis by FFT and the spectrum is shifted and sorted to obtain the spectrum; for each pixel point, coherent accumulation is performed in azimuth and the cross-resolution unit distance motion correction processing is performed, and finally an arc SAR image of the monitoring area is obtained. The fast BP algorithm in the present invention has a low amount of calculation and storage, and has good adaptability in the case of large-distance motion. Combined with the rotating scanning SAR, it can cover a large range of 360-degree surrounding scenes in one observation, and has high efficiency and wider application.

Description

Technical Field

[0001] The invention belongs to the technical field of radar signal processing, and in particular relates to a high-precision imaging solution of a rotating scanning ground-based SAR based on a fast BP algorithm. Background Art

[0002] Synthetic aperture radar (SAR) is a high-resolution radar that uses pulse compression technology in the range direction and synthetic aperture technology in the azimuth direction to obtain high-resolution images of the target. It can observe the earth all day and all weather, and can penetrate the surface and vegetation at a specific frequency to obtain information under the surface. Therefore, it has a wide range of applications in geological surveys, remote sensing imaging and deformation detection, target recognition and other fields. Ground-based synthetic aperture radar (GB-SAR) combines the SAR imaging principle with electromagnetic wave interference technology, and uses the system parameters, attitude parameters and geometric relationship between the orbits of the sensor to accurately measure the spatial position and tiny changes of a certain point on the surface, and can detect millimeter-level or even sub-millimeter-level surface deformation. It has high monitoring accuracy and short revisit cycle, and can realize continuous and repeated observation of a fixed area, real-time feedback of monitoring information, and relatively low maintenance costs.

[0003] The radar antenna of the traditional GBSAR system moves along a linear track to scan the monitoring scene. This system has exposed certain technical limitations in practical applications, mainly manifested in a small field of view, which cannot cover a large range of surrounding scenes in one observation. A new GBSAR observation mode, the rotating scanning ground-based SAR (Arc SAR) system, came into being. The synthetic aperture of the rotating scanning ground-based SAR is generated by rotating the antenna attached to the end of the rotating arm. It uses the rotation direction of the radar platform as the azimuth direction and the radial direction of the platform rotation direction as the range direction. Under the premise of ensuring high-resolution imaging, the rotating scanning ground-based SAR system can cover a large range of surrounding 360-degree scenes in one observation, which effectively improves the field of view of the radar system.

[0004] Although the feasibility and unique advantages of Arc SAR have been verified in recent years, there are still problems to be solved in its imaging algorithm. The Arc SAR radar antenna scans the scene along an arc trajectory. This scanning method increases the difficulty of imaging. Therefore, it is necessary to study the imaging algorithm of the rotating scanning ground-based SAR system.

[0005] The backprojection (BP) algorithm is an accurate time-domain algorithm, which is essentially a beamforming of the antenna phase array along a specific direction. In the range compression-azimuth time domain, the BP algorithm reconstructs each pixel on the imaging grid along the slant range history, which ideally solves the range-azimuth coupling problem. The BP algorithm has the ability of accurate motion compensation and is suitable for image focusing under imaging conditions such as arbitrary tracks, arbitrary bandwidths and accumulation angles. It does not have the plane wavefront assumption, has no restrictions on the range of the imaging scene, and the focused image has no geometric distortion, so the BP algorithm does not require space-variant filtering after image focusing and interpolation-based range curvature correction. However, compared with other algorithms, the computational complexity of the BP algorithm is extremely large, which greatly limits the real-time application of the BP algorithm in large-scale imaging scenarios. Therefore, it is of great significance to study how to reduce the computational complexity of the BP algorithm. Summary of the invention

[0006] In order to solve the problems existing in the above technologies, the present invention provides a high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm. The problem to be solved by the present invention is achieved through the following technical solutions:

[0007] A high-precision imaging scheme for rotating scanning ground-based SAR based on a fast BP algorithm, characterized by comprising the following steps:

[0008] (1) Based on the gridded high-grid auxiliary data H(m s ,n) Calculate the azimuth phase history function A(m s ,n;m')=exp[jΦ(m s ,n;m')], the number of pulse cycles m0(m s ,n), and are stored in the data file in the form of three-dimensional arrays;

[0009] (2) Recording the echo signal sampling data y(m, k) of the monitoring area through arc rotation scanning;

[0010] (3) For each modulation period m, the data {y(m,k)|k=0,1,…,K-1} is subjected to spectrum analysis by FFT, and the spectrum is shifted and sorted to obtain the corresponding spectrum {Y(m,n)|n=0,1,…,K'-1};

[0011] (4) For each pixel (m s ,n), perform correlation accumulation in the azimuth direction and perform motion correction and sorting across the resolution unit distance, namely:

[0012]

[0013] m s=M',M'+1,...,M-M'-1; n=0,1,...,round[(K' / K)(R max -R min ) / ΔR s ]-1 obtain the arc SAR image of the corresponding monitoring area.

[0014] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art:

[0015] 1. There is no need to lay out a long platform motion track on the ground, and a short arc motion trajectory can be used to scan a large circular monitoring area.

[0016] 2. Allows the radar to have a larger beam width in azimuth and achieve higher angular resolution with a relatively short rotating arm length.

[0017] 3. It can realize tomographic three-dimensional imaging, which is convenient for combining deformation displacement with three-dimensional terrain and terrain slope, making deformation monitoring intuitive and visualized.

[0018] 4. Compared with the original BP imaging algorithm, it significantly reduces the amount of calculation and storage, and is more adaptable to large-distance movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a rotation scanning ground-based SAR imaging method based on a fast BP algorithm provided by an embodiment of the present invention.

[0020] Figure 2 A rotating scanning ground-based SAR signal model is provided in an embodiment of the present invention.

[0021] In the signal model, it is assumed that the shortest distance from a scatterer T to the arc in the monitoring area is R s , the nearest point is S, the rotation center (fulcrum, the center of the arc) is O, the starting point of the arc is P, and the angle between OS and OP (the rotation angle at S) is θs; then the projection point T1 of the scatterer on the arc plane is on the extension line of OS, ST1 = R 1s ; Line segment TT1 is perpendicular to the arc plane, TT1 = H s . DETAILED DESCRIPTION

[0022] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0023] The basic idea of ​​the fast BP algorithm is to approximate the signal model within each modulation period so that the processing in the radial distance direction can be converted into FFT processing; and according to the fact that the distance migration of each scattering point in the synthetic aperture length of the arc SAR is small (much smaller than that of the linear SAR), generally less than 2 radial distance resolution unit widths (to be analyzed later), the distance migration correction for each pixel point only needs to be performed at the close range R of the pixel point. s It is performed on the range channel where the information is located and the two adjacent range channels, without the need for frequency domain range migration correction or large-scale cross-range resolution unit accumulation in the time domain, making the azimuth accumulation processing easier to achieve.

[0024] According to the attached figure, the azimuth phase history function and the number of pulse cycles required for the movement across the resolution unit are calculated in advance based on the gridded elevation auxiliary data and stored in the data file in the form of a three-dimensional array; secondly, the echo signal sampling data of the monitoring area is recorded by arc rotation scanning. For each modulation period m, the data is spectrally analyzed by FFT and the spectrum is shifted and sorted to obtain the spectrum. For each pixel point, coherent accumulation is performed in azimuth and the cross-resolution unit distance motion correction processing is performed to obtain the arc SAR image of the monitoring area.

Claims

1. A high-precision imaging scheme for rotating scanning ground-based SAR based on a fast BP algorithm, characterized in that Follow these steps: (1) Based on the gridded high-grid auxiliary data H(m s ,n) Calculate the azimuth phase history function A(m s ,n;m')=exp[jΦ(m s ,n;m')], the number of pulse cycles m0(m s ,n), and are stored in the data file in the form of three-dimensional arrays; (2) Recording the echo signal sampling data y(m, k) of the monitoring area through arc rotation scanning; (3) For each modulation period m, the data {y(m,k)|k=0,1,…,K-1} is subjected to spectrum analysis by FFT, and the spectrum is shifted and sorted to obtain the corresponding spectrum {Y(m,n)|n=0,1,…,K'-1}; (4) For each pixel (m s ,n), perform correlation accumulation in the azimuth direction and perform motion correction and sorting across the resolution unit distance, namely: m s =M',M'+1,...,M-M'-1;n=0,1,...,round[(K' / K)(R max -R min ) / ΔR s ]-1 The arc SAR image of the corresponding monitoring area is obtained.

2. A high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm according to claim 1, characterized in that: The azimuth phase history function and the number of pulse cycles required for cross-resolution unit movement are calculated in advance based on the gridded elevation auxiliary data.

3. A high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm according to claim 1, characterized in that: The number of pulse cycles required for the cross-resolution unit movement is stored in the data file in the form of a three-dimensional array.

4. A high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm according to claim 1, characterized in that: The signal model in each modulation period is first approximated so that the processing in the radial distance direction can be converted into FFT processing.

5. A high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm according to claim 1, characterized in that: The distance migration correction for each pixel only requires the near distance R of the pixel to be s It is performed on the range channel where the information is located and the two adjacent range channels, without the need for frequency domain range migration correction or large-scale cross-range resolution unit accumulation in the time domain, making the azimuth accumulation processing easier to achieve.

6. A high-precision imaging solution for rotating scanning ground-based SAR based on a fast BP algorithm according to claim 1, characterized in that: The data comes from the echo signal sampling data of the monitoring area recorded by arc rotation scanning.

Images