A fast data preprocessing method based on navigation satellite bistatic InSAR

Abstract

The application discloses a kind of based on navigation star bistatic InSAR's fast data preprocessing method, on the premise of guaranteeing the amplitude and phase accuracy of imaging result, every 1ms echo data is truncated into 50 portions 20us data, and the 20us data corresponding to 1ms reference signal is cross-correlated;50 results obtained by calculation are accumulated again, and the sum obtained is the result of pulse compression;Finally, the final image is obtained using the result of the outcome. The 20us data pulse compression method used changes the size of PRT in the imaging process to speed up the pulse compression speed, which can reduce the difficulty of data processing and the physical cost required for storage under the premise of ensuring the accuracy of imaging results.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of bistatic synthetic aperture radar, and particularly relates to a fast data preprocessing method based on navigation satellite bistatic InSAR. BACKGROUND

[0002] GNSS-InSAR (Differential Interferometric SAR Based on Global Navigation Satellite System) is a typical application of air-ground bistatic SAR, which uses on-orbit navigation satellites as external radiation sources and near-ground static receiving mechanisms to form a bistatic SAR system, and has the advantages of high data rate, low cost and easy deployment. At the same time, navigation satellite constellation resources are rich, such as Beidou, GPS, GLONASS, Galileo, etc. in orbit, and have the characteristics of global coverage and short revisit time, which can provide multi-angle data sets and rich space-time resources.

[0003] In order to achieve better azimuth resolution, the single imaging time of navigation satellite is about 600s. The single imaging data is huge, which causes great pressure on data transmission and limits the real-time processing and rapid deformation monitoring reaction ability of the system. SUMMARY

[0004] Therefore, the application provides a fast data preprocessing method based on navigation satellite bistatic InSAR, which can reduce the difficulty of data processing and the physical cost required for storage under the premise of ensuring the accuracy of imaging results.

[0005] To achieve the above purpose, the fast data preprocessing method based on navigation satellite bistatic InSAR includes the following steps:

[0006] Step one, calculating the length of a single processing data segment according to the sampling rate of the navigation satellite signal;

[0007] Step two, truncating the echo according to the calculated data segment length;

[0008] Step three, truncating the echo corresponding to the reference signal and zero-padded to be consistent with the echo;

[0009] Step four, pulse compression processing is performed on the truncated echo and reference signal, and the obtained results are accumulated;

[0010] Step five, using the accumulated results to perform a backward projection algorithm to obtain the final image.

[0011] In the step one, the length of the single processing data segment is 2 raised to the power of n, and is greater than twice of the data amount of 10us under the current signal sampling rate.

[0012] In the step two, the signal is sequentially truncated into 50 data segments with 10us according to the data segment length in the step one.

[0013] In the step three, the reference signal data is sequentially truncated with 10us, and is zero-padded to be consistent with the echo signal format.

[0014] In the step five, the accumulated result is taken as the pulse compression result of the whole 1ms for the back-projection algorithm imaging.

[0015] Advantageous effects

[0016] 1. The method mainly aims at the characteristics of the navigation satellite bistatic imaging algorithm occupying a large amount of memory and being slow, through truncating the navigation satellite echo data in the pulse compression process of data processing, correlating the corresponding data in the reference signal, accumulating the results, obtaining the accumulation, taking the accumulation as the pulse compression result, and then using the pulse compression result to calculate the phase of the imaging plane to obtain the final image.

[0017] 2. The 20us data pulse compression method adopted in the application only needs to perform 50 times of Fourier transform of one-fiftieth data amount when performing Fourier transform on the echo signal and the reference signal, and the operation amount is reduced by 44%. After being converted to the time domain after the conjugate multiplication, only 50 times of inverse Fourier transform of one-hundredth data amount is still needed, and the operation amount is also reduced by 44%. At the same time, compared with the PRT of 1ms, the physical memory required for storing the pulse compression result is saved by 98%.

[0018] 3. In the application, the size of the PRT in the imaging process is changed to speed up the pulse compression, which can reduce the difficulty of data processing and the physical cost required for storage under the premise of ensuring the accuracy of the imaging result. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 It is a system configuration diagram of the embodiment of the application;

[0020] Figure 2 It is a system equipment diagram of the embodiment of the application;

[0021] Figure 3 It is a simulation imaging result diagram without using the application;

[0022] Figure 4 It is a simulation imaging result diagram using the application;

[0023] Figure 5 for using the simulation imaging result map of the present application;

[0024] Figure 6 for using the simulation imaging result map of the present application. DETAILED DESCRIPTION

[0025] The present application is described in detail below with reference to the accompanying drawings and embodiments.

[0026] The present application is a fast data preprocessing method based on navigation satellite bistatic InSAR, comprising the following steps:

[0027] Step one, calculate the length of single processing data segment according to the navigation satellite signal sampling rate, specifically:

[0028] Suppose the navigation satellite signal sampling rate is 31MHz, then the data amount of 10us is 310, in order to make 310 points fully pulse compressed, the single processing data segment needs to be above 310*2, in order to make subsequent FFT operation convenient, here take the minimum 2x power above 310*2, here take 1024.

[0029] Step two, truncate the echo according to the calculated data segment length, specifically:

[0030] Take 1ms navigation satellite collected signal, then the signal contains data amount of 31000, according to the order of 1-1024, 620+1-620+1024, 620*2+1-620*3, truncate and reorganize the signal into the first 49 rows of 50*1024 matrix, and use 31000-1024-31000 data for the 50th row.

[0031] Step three, truncate the echo corresponding to the reference signal, and zero fill to the length equal to the echo matrix, specifically:

[0032] Take 1ms navigation satellite reference signal, then the signal contains data amount of 31000, according to the order of 1-620, 620+1-620*2, 620*2+1-620*3, truncate the reference signal and all zero fill to the data segment with length of 1024, and finally reorganize into a 50*1024 matrix.

[0033] Step four, pulse compression processing is performed on the truncated data, and accumulation is performed to obtain the final result, specifically:

[0034] Take the reorganized matrix of the echo and the reorganized matrix of the reference signal to perform pulse compression processing, and obtain a 50*1024 result matrix, accumulate the matrix by row, and the obtained 1*1024 result is regarded as the pulse compression result of the whole 1ms echo.

[0035] Step five, using the pulse compression result, backward projection imaging is carried out to obtain the final imaging result, specifically:

[0036] The imaging area is divided into a grid, the distance between each point on the grid and the bistatic radar is calculated, the pulse compression result is searched, and the final imaging image is obtained after phase compensation and accumulation.

[0037] Simulation experiment verification:

[0038] In this simulation experiment, according to the parameters in Table 1, a point target with coordinates (100, 100) is simulated in a 1200x1200 grid scene.

[0039] Table 1 Simulation parameter table

[0040]

[0041]

[0042] The operation amount of PRT for 1ms, 10us in each process of pulse compression is shown in Table 2 and Table 3.

[0043] Table 2 Operation amount of pulse compression process when PRT is 1ms

[0044] Processing steps Complex multiplication Values Matched signal N 3.1e4 Matched, echo signal FFT N log2(N) 4.63e5 Conjugate multiplication N 3.1e4 Post-multiplication spectrum IFFT N / 2 log2(N) 2.31e5 Total 7.56e5

[0045] Table 3 Operation amount of pulse compression process when PRT is 10us

[0046] Processing steps Complex multiplication Values Matched signal N 3.1e4 Matched, echo signal FFT N log2(N / 100) 2.57e5 Conjugate multiplication N 3.1e4 Post-multiplication spectrum IFFT N / 2 log2(N / 100) 1.28e5 Total 4.47e5

[0047] Compared with PRT of 1ms, the operation amount of PRT of 10us in the pulse compression process is reduced by 41%, and the physical memory occupied by the pulse compression result is reduced by 99%. The simulation result of PRT of 1ms is shown in Figure 3 , and the simulation result of PRT of 10us is shown in Figure 4 . The points within the range of 3db maximum value of the image are shown in Figure 5 , 6 . It can be seen that the range resolution and peak value are consistent. From the result of subtracting the amplitude difference in db, the maximum difference is at the position of the sidelobe, which is 132.5dB. From the imaging result, it can be seen that the main lobe peak value is 171dB, which is about 40dB different, and will not affect the imaging effect.

[0048] The experiment verifies that the ecological park in Changshu City is selected as the imaging area, the experimental time is July 10, 2021, and the data received and processed comes from Beidou PRN 9 satellite. The scattering points in the results of PRT of 1ms and 10us are basically the same, and the maximum value is 220.7.

[0049] The above specific embodiments only describe the design principles of the present application, and the shapes and names of the components in the description can be different and are not limited. Therefore, those skilled in the art of the present application can modify or equivalently replace the technical solutions described in the foregoing embodiments; and these modifications and replacements do not deviate from the purpose and technical solutions of the present application, and should all belong to the protection scope of the present application.

Claims

1. A fast data preprocessing method based on navigation star bistatic InSAR, characterized in that, Comprising the following steps: Step one, according to the navigation satellite signal sampling rate calculation length of single processing data segment, specifically: Suppose the navigation satellite signal sampling rate is 31MHz, then the data amount of 10us is 310, in order to make 310 points fully pulse compression, single processing data segment needs to be above 310*2, in order to make subsequent FFT operation convenient, here take 310*2 above the minimum power of 2, here take 1024; Step two, according to the calculated data segment length, the echo is truncated, specifically: Take 1ms navigation satellite acquisition signal, then the signal contains data amount is 31000, according to 1-1024, 620+1-620+1024, 620*2+1-620*2+1024 order to the signal is truncated and reorganized into 50*() matrix of the first 49 rows of data, the 50th row uses 31000-1024-31000 data; Step three, the reference signal is truncated according to the echo, and zero is filled to be equal to the length of the echo matrix, specifically: Take 1ms navigation satellite reference signal, then the signal contains data amount is 31000, according to 1-620, 620+1-620*2, 620*2+1-620*3 order to the reference signal is truncated and all zero to the length of 1024 data segment, finally reorganized into 50*1024 matrix; Step four, the truncated echo and reference signal are pulse compressed and accumulated to get the final result, specifically: Take the reorganized matrix of echo and the reorganized matrix of reference signal for pulse compression processing, get a 50*1024 result matrix, accumulate the matrix according to the row, the 1*1024 result is regarded as the pulse compression result of 1ms echo as a whole; Step five, using the pulse compression result, backward projection imaging is carried out to get the final imaging result, specifically: Divide the imaging area into grid, calculate the distance between each point on the grid and the bistatic radar, search the pulse compression result, then accumulate after phase compensation to get the final imaging image.

Images