An ultra-high-resolution FMCW system spaceborne sliding spotlight mode SAR imaging method

Abstract

This invention discloses an ultra-high resolution FMCW-based spaceborne sliding spotting mode SAR imaging method, which receives the reflected signals of sub-blocks in the region to be imaged; performs Fourier transform on the fast time portion of the reflected signal and then phase compensation on the slow time portion, converting the resulting signal into a virtual equivalent direct sampling matched filter echo signal; performs range pre-filtering; and performs frequency domain f i The domain flip translation is the translation amount under a unified reference slant range; fast time domain sign and scale replacement is performed; the range time domain is converted to the frequency domain and downsampled; the downsampled signal is imaged to obtain sub-block images; several sub-block images are fused to obtain the SAR image of the area to be imaged; by using a spaceborne ultra-high resolution SAR fast imaging method based on range-to-frequency domain downsampling, the computational load of block imaging processing can be greatly reduced.

Description

Technical Field

[0001] This invention belongs to the field of microwave remote sensing technology, and in particular relates to an ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method. Background Technology

[0002] Strip mode is the most basic operating mode of spaceborne SAR. Its advantage is that it can continuously image large areas. However, due to limitations such as antenna gain, the azimuth resolution of the radar cannot be arbitrarily increased as the antenna length decreases. Its azimuth resolution cannot be higher than half the size of the azimuth antenna.

[0003] In spotting mode, as the platform moves, the radar beam is controlled to consistently illuminate a small area of ​​the ground by adjusting its direction. This beam pointing control improves the synthetic aperture time of the target, thus achieving higher azimuth resolution. The drawbacks are a smaller azimuth imaging area and the inability to perform continuous imaging.

[0004] The sliding spotting mode, a more general approach between strip and spotting modes, offers higher spatial resolution than the traditional strip mode and a wider azimuth mapping band than the spotting mode. Spotting and strip modes can be considered special cases of the sliding spotting mode. Spotting mode occurs when the irradiated area moves at zero speed across the ground; striping mode occurs when the irradiated area moves at the same speed as the payload platform.

[0005] Traditional sliding spotlight SAR often employs a pulsed emission system; serial processing of full-aperture full-frame image echo data; unified reference slant range and unified coordinate system algorithms; and imaging geometry models based on straight-line orbits.

[0006] Traditional X-band and C-band spaceborne sliding beam high-resolution SAR often employs a block imaging processing method when the azimuth swath is relatively small. However, when entering high-frequency bands (such as Ka-band), due to the higher frequency, when the system pulse repetition frequency (PRF) is fixed, the maximum Doppler bandwidth that the imaging processing can tolerate will be much lower than that of the X-band or C-band. Consequently, the instantaneous illumination range of the antenna will also be much lower than that of the X-band or C-band.

[0007] Therefore, for high-frequency band (such as Ka-band) spaceborne sliding spotting high-resolution SAR imaging requiring a large azimuth swath, the traditional block imaging processing method results in a huge number of azimuth points, leading to a sharp increase in computational load. On the other hand, in spaceborne ultra-high resolution sliding spotting SAR imaging, due to the long synthetic aperture length, the traditional straight-track approximation model has a large error. In order to reduce the spatial variation error caused by track curvature, block processing is also required in the range direction. Summary of the Invention

[0008] The purpose of this invention is to provide an ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method to reduce the computational load of imaging processing.

[0009] This invention adopts the following technical solution: a spaceborne sliding spotting mode SAR imaging method based on ultra-high resolution FMCW system, comprising the following steps:

[0010] Receive the reflected signals from sub-blocks of the region to be imaged;

[0011] The fast time portion of the reflected signal is subjected to Fourier transform, and then the slow time portion is phase compensated to obtain signal S. if_com1 (f i ,t m ); where f i Represents frequency, t m Indicates slow time;

[0012] S if_com1 (f i ,t m The echo signal S is converted into a virtual equivalent direct sampling matched filter. if_com2 (f i ,t m );

[0013] For S if_com2 (f i ,t m Perform range pre-filtering to obtain S if_com3 (f i ,t m );

[0014] S if_com3 (f i ,t m The target in the frequency domain f i The domain flip translation is the translation amount under a unified reference slant range, resulting in S. if_com4 (f i ,t m );

[0015] S if_com4 (f i ,t m (This is obtained by performing fast time domain sign and scaling substitution) in, Indicates a fast time;

[0016] Will The distance is obtained by frequency conversion in the time domain to obtain s(ω,t) m ), and for s(ω,t mDownsampling is performed in the ω-domain; where ω represents the angular frequency domain;

[0017] The downsampled signal is imaged to obtain a sub-block image;

[0018] By fusing several sub-block images, a SAR image of the region to be imaged is obtained.

[0019] Furthermore, before receiving the reflected signal from a sub-block of the region to be imaged, the following steps are also included:

[0020] Calculate the distance deviation between the circular orbit and the approximately equivalent straight orbit of the SAR antenna at different apertures and different imaging points;

[0021] The imaging area of ​​a sub-block is determined when the distance deviation within the imaging area of ​​the sub-block is not greater than the threshold.

[0022] Furthermore, after receiving the reflected signal from the sub-block of the region to be imaged, the process also includes:

[0023] Perform overall orbital curvature compensation on the imaging area of ​​the sub-block.

[0024] Furthermore, after performing overall orbital curvature compensation on the imaging region of the sub-block, the process also includes:

[0025] The signal after overall track curvature compensation is received by variable reference slant range de-slant reception.

[0026] Furthermore, regarding S if_com2 (f i ,t m Perform range pre-filtering to obtain S if_com3 (f i ,t m )include:

[0027] S if_com3 (f i ,t m ) = S if_com2 (f i ,t m )H filter (f i ),

[0028] Among them, H filter (f i ) is the filtering function. The filtering function retains the data for the distance gate where the sub-block is located, and sets the other distance gates to zero.

[0029] Furthermore, The distance in the time domain and frequency conversion domain includes:

[0030]

[0031] in, Indicates to exist Perform a Fourier transform on the domain.

[0032] Furthermore, the variable reference slant range de-slant reception of the signal after overall track curvature compensation includes:

[0033]

[0034] in, This indicates the signal after overall track curvature compensation. This represents the reference signal corresponding to the i-th reference slant distance.

[0035] Furthermore, S if_com1 (f i ,t m The echo signal S is converted into a virtual equivalent direct sampling matched filter. if_com2 (f i ,t m )include:

[0036]

[0037] Where c represents the speed of light, f c R represents the carrier frequency of the antenna's transmitted signal. Δ R represents the distance difference between the antenna at the azimuth sampling position and the target, and the reference distance corresponding to the reference signal. Δi =R t -R ref_i R t R represents the distance from the target to the antenna. ref_i This is the i-th reference slope distance.

[0038] Furthermore, S if_com4 (f i ,t m Specifically:

[0039]

[0040] Where A represents amplitude, T p γ represents the radar transmit pulse width, and γ represents the antenna frequency modulation slope.

[0041] The beneficial effects of this invention are: by using a spaceborne ultra-high resolution SAR fast imaging method based on range-direction frequency domain downsampling, under the preprocessing framework of variable reference slant range and variable coordinate system, the method first performs overall pulse compression in the range direction, then divides the image into blocks in the pulse compression domain, and then performs range-direction time domain pre-filtering, and finally performs range-direction frequency domain downsampling, which can significantly reduce the computational load of block imaging processing. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of a spaceborne sliding spotting mode SAR imaging method based on an embodiment of the present invention using an ultra-high resolution FMCW system.

[0043] Figure 2 This is a schematic diagram of the instantaneous illumination geometry for sliding beam imaging in an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the sliding bundle geometry in an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the sub-aperture decomposition of the sliding beam imaging azimuth sub-block in an embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the sub-aperture decomposition of the sliding beam imaging azimuth sub-block in an embodiment of the present invention;

[0047] Figure 6 This is a flowchart of the parallel fast imaging processing method for Ka-band high-resolution wide-swath SAR in an embodiment of the present invention.

[0048] Figure 7 This is a schematic diagram of strabismus geometric imaging in an embodiment of the present invention;

[0049] Figure 8 This is a schematic diagram of the distance migration curve after distance compression under the oblique view situation in an embodiment of the present invention;

[0050] Figure 9 This is a schematic diagram of variable reference coordinate system imaging in an embodiment of the present invention;

[0051] Figure 10 This is a schematic diagram of the distance migration curve after distance compression under the variable reference coordinate system (equivalent distance movement correction) in the case of oblique view in an embodiment of the present invention. Detailed Implementation

[0052] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0053] This invention proposes a parallel and fast imaging processing method for high-frequency band (such as Ka-band) spaceborne high-resolution wide-swath SAR. By establishing the geometric relationship between wide-swath and high-resolution imaging, designing the instantaneous illumination range of the antenna, and then constructing independent imaging blocks in the azimuth and range directions, parallel and fast imaging is finally completed on the GPU.

[0054] In spaceborne sliding spotlight SAR, traditional pulse transmission systems require high pulse power and are costly. By designing an FMCW (Fluidized Motion Wave) system, peak power and cost can be reduced. Traditional pulse-based high-resolution (e.g., 0.5m) sliding spotlight SAR imaging typically uses direct sampling reception. However, when moving to ultra-high resolution (e.g., 0.05m) sliding spotlight SAR imaging, using an FMCW deskewing (beat) reception system can reduce the large amount of sampled data caused by direct sampling reception under wide bandwidth.

[0055] This invention proposes a virtual equivalent direct sampling method for FMCW deslant (beat) reception systems. By establishing the correlation transformation relationship between the echo signal after direct sampling and the signal after deslant reception, FMCW deslant reception data can be processed based on the imaging processing method of direct sampling.

[0056] In addition, during spaceborne ultra-high resolution deslant reception SAR imaging, due to the high resolution and long synthetic aperture length, if the traditional unified reference slant range processing method is used, the range sampling bandwidth will increase sharply, and the sampling rate will increase sharply. Furthermore, under large slant view, the data matrix caused by ultra-high resolution range migration will increase sharply in the range direction.

[0057] This invention proposes a large sampling data dimensionality reduction processing technique for spaceborne large oblique angle ultra-high resolution SAR imaging. By combining a variable coordinate system processing method with a variable reference slant range sampling data dimensionality reduction preprocessing method, the amount of sampling data can be significantly reduced.

[0058] In addition, for high-frequency ultra-high resolution imaging, the shorter wavelength results in a lower tolerance for range blocks. Furthermore, the longer synthetic aperture length corresponding to high-frequency ultra-high resolution also results in a lower tolerance for range blocks. Therefore, it is necessary to increase the number of range blocks in the echo data domain, but this also brings another problem: an increase in the overall computational load of range data.

[0059] This invention proposes a fast ultra-high resolution SAR imaging method based on range-direction frequency domain downsampling. Under the preprocessing framework of variable reference slant range and variable coordinate system, the method first performs overall pulse compression in the range direction, then divides the image into blocks in the pulse compression domain, and finally performs range-direction time domain pre-filtering, followed by range-direction frequency domain downsampling. This method can significantly reduce the computational load of block imaging processing.

[0060] This invention discloses an ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method, such as... Figure 1 As shown, the process includes the following steps: receiving the reflected signal from a sub-block of the region to be imaged; performing a Fourier transform on the fast time portion of the reflected signal and then performing phase compensation on the slow time portion to obtain signal S. if_com1 (f i ,t m); where f i Represents frequency, t m Indicates slow time; S if_com1 (f i ,t m The echo signal S is converted into a virtual equivalent direct sampling matched filter. if_com2 (f i ,t m ); for S if_com2 (f i ,t m Perform range pre-filtering to obtain S if_com3 (f i ,t m ); S if_com3 (f i ,t m The target in the frequency domain f i The domain flip translation is the translation amount under a unified reference slant range, resulting in S. if_com4 (f i ,t m ); S if_com4 (f i ,t m (This is obtained by performing fast time domain sign and scaling substitution) in, Indicates a fast time; will The distance is obtained by frequency conversion in the time domain to obtain s(ω,t) m ), and for s(ω,t m Downsampling is performed in the ω domain; where ω represents the angular frequency domain; the downsampled signal is imaged to obtain sub-block images; several sub-block images are fused to obtain a SAR image of the area to be imaged.

[0061] By employing a spaceborne ultra-high resolution SAR fast imaging method based on range-direction frequency domain downsampling, under the preprocessing framework of variable reference slant range and variable coordinate system, the method first performs overall pulse compression in the range direction, then divides the image into blocks in the pulse compression domain, and finally performs range-direction time domain pre-filtering, followed by downsampling in the range-direction frequency domain. This significantly reduces the computational load of block imaging processing.

[0062] In SAR imaging, to improve azimuth resolution, a sliding spotting mode can be used, which synthesizes a longer aperture by rotating the antenna angle to focus imaging on a specific area. For example... Figure 2 The diagram shows the instantaneous illumination geometry for sliding focus imaging. In this mode, a longer synthetic aperture produces a larger Doppler bandwidth, which can cause azimuth Doppler blurring if a strip mode PRF is used. To address this issue, azimuth compression (de-Doppler blurring) processing is required before imaging. Although this preprocessing method effectively reduces the PRF requirement, it still has a minimum PRF requirement, specifically constrained by the following formula:

[0063]

[0064] Where, Δ uc R is the spatial sampling step size of the original data in the azimuth direction. c Y0 is the slant distance of the center point of the instantaneously illuminated area for imaging, Y0 is the azimuth swath width of the instantaneously illuminated area for imaging, λ represents the wavelength of the electromagnetic wave, and v0 is the wavelength of the electromagnetic wave. s This indicates the speed at which the satellite moves, and PRF stands for Pulse Repetition Frequency.

[0065] From Formula 1, we can obtain:

[0066]

[0067] Therefore, in the sliding clustering mode, the minimum PRF is:

[0068]

[0069] Where, λ min This is the minimum wavelength within the electromagnetic wave bandwidth range. Conversely, when the PRF is fixed, the maximum azimuth swath width of the instantaneously illuminated imaging area is:

[0070]

[0071] When the instantaneous illumination swath of the antenna is equal to the total imaging swath in the azimuth direction, it can be processed using a whole-block imaging method. However, when entering high-frequency bands such as the Ka band, due to the higher frequency, the maximum Doppler bandwidth that the imaging processing can tolerate when the system PRF is fixed will be much lower than that of the X-band or C-band, and consequently, the instantaneous illumination range of the antenna will also be much lower than that of the X-band or C-band.

[0072] For the Ka band, with an orbital altitude of 800 km, an angle of view of 30°, a scene center slant distance of 923.76 km, a carrier frequency of 35 GHz (wavelength 0.0086 m), a PRF of 3000 Hz, and a satellite velocity of 7617 m / s, the maximum allowable azimuth swath width for instantaneous imaging, calculated according to Equation 4, is 1.5453 km (±0.7726 km). If the azimuth swath width requirement is 15 km to 20 km, the PRF needs to be increased to increase the instantaneous illumination range. Calculations show that for a 15 km azimuth swath width, a PRF of 30000 Hz is required, but a higher PRF would increase the design complexity of the system.

[0073] If a lower PRF is used, according to the sliding beam geometry, the ratio of the total azimuth width to the instantaneous illumination range is relatively large, resulting in a longer overall synthetic aperture length due to the total azimuth width. Therefore, the total number of sampling points in the azimuth increases dramatically. If the traditional whole-block imaging processing method is used, the computational load increases dramatically.

[0074] To address this challenge, this invention proposes a parallel fast imaging processing method for high-frequency spaceborne high-resolution wide-swath SAR. This method establishes the geometric relationship between the wide-swath and high-resolution imaging, designs the instantaneous illumination range of the antenna, constructs azimuth-independent imaging blocks, and finally completes parallel fast imaging on a GPU. The scheme is described in detail below.

[0075] For Equation 4, when the antenna azimuth dimension is fixed, the maximum azimuth swath width of the instantaneous imaging illumination area is:

[0076]

[0077] Where D represents the aperture of the antenna.

[0078] Considering that the sliding beamforming mode, by gradually sliding across the aforementioned footprints and illuminating different footprints with different apertures, can ultimately achieve a larger azimuth swath than a single footprint, when the azimuth resolution is determined, the effective synthetic aperture length for each imaging point (referring to the imaging pixel position) is:

[0079]

[0080] Where, r A This refers to the azimuth resolution. According to Equation 6, to ensure that the azimuth resolution of each imaging point meets the requirements, such as... Figure 3 As shown, the total synthetic aperture length of the entire imaging region is:

[0081]

[0082] Where L is the total synthetic aperture length of the entire imaging region, L0 is the effective synthetic aperture length of each imaging point, Y is the total swath width of the entire imaging region, and v s v is the speed at which the satellite moves. f The speed at which the footprints moved.

[0083] When entering high-frequency bands such as the Ka band, as calculated above, the ratio of the total azimuth swath to the instantaneous illumination range is relatively large, resulting in a longer overall synthetic aperture length due to the total azimuth swath. Therefore, the total number of sampling points in the azimuth increases dramatically. If the traditional whole-block imaging processing method is used, the computational load increases dramatically.

[0084] like Figure 4As shown, considering the relative independence of each imaging footprint during the Ka-band wide-swath sliding process, and the relative independence between each imaging sub-aperture corresponding to each imaging footprint, the entire imaging area can be decomposed into independent sub-blocks and sub-apertures for imaging. The number of sampling points for each independent sub-block and sub-aperture will be greatly reduced, thus significantly reducing the memory and computational load of each imaging sub-block. At the same time, due to the relative independence between each imaging sub-block, the parallel architecture of the GPU is used to process each imaging sub-block in parallel, and finally the stitching is completed in the image domain.

[0085] Due to the relative independence of each imaging sub-block, stitching is ultimately performed in the image domain after each sub-block is imaged. This provides a higher error tolerance than stitching based on coherent superposition in the signal domain, making image domain stitching easier to implement. Furthermore, block imaging can also reduce the influence of spatial variability in the imaging region, improving imaging accuracy.

[0086] In spaceborne ultra-high resolution sliding spotlight SAR imaging, due to the long synthetic aperture length, the influence of orbit curvature cannot be ignored. Directly approximating this with a straight orbit model would cause significant errors. Therefore, this can be addressed by dividing the imaging into blocks along the range direction and compensating for the orbit curvature error in each sub-block. For example... Figure 5 The image shows a schematic diagram of a circular curved track.

[0087] Let the antenna sampling coordinates under the circular orbit be (X1, Y1, Z1), and the corresponding antenna sampling coordinates under the approximately straight orbit be (X0, Y0, Z0). The X direction is the direction of the antenna's projection onto the ground towards the area to be imaged, the Y direction is the antenna's direction of travel (i.e., the satellite's direction of motion), and the Z direction is the direction of the antenna's projection onto the ground towards the antenna. Using the height of the imaging plane as the zero-point coordinate in the Z direction, and the midpoint of the antenna within the synthetic aperture as the zero-point coordinates in the X and Y directions, we have:

[0088]

[0089] Where H is the satellite's altitude, R is the Earth's radius, and O is the Earth's center. look_angle is the antenna's angle of view, L is half the effective synthetic aperture length, and theta_curve is the geometric angle between the circular orbit and the Earth's center when the aperture is at point L. theta Let be the geometric angle between the approximate straight track and the center of the Earth. Let the coordinates of the imaging point be (x, y1, 0). Then, according to the calculation formula for the three-dimensional coordinate system, the distance deviation (i.e., phase difference) between the circular track and the approximate equivalent straight track at different apertures and different imaging point positions can be calculated.

[0090] In other words, before receiving the reflected signal from a sub-block of the area to be imaged, the distance deviation between the circular orbit and the approximately equivalent straight orbit of the SAR antenna at different apertures and different imaging points is calculated. When the distance deviation within the imaging area of ​​the sub-block is not greater than a threshold (e.g., not exceeding 45°), i.e., spatial variation can be ignored, the imaging area of ​​the sub-block is determined. Then, overall orbit curvature compensation can be performed on the imaging area of ​​the sub-block. During compensation, the distance deviation value of the center point of the imaging area of ​​the sub-block is used as a reference to compensate for all positions in the imaging area of ​​the sub-block.

[0091] Based on the above principles, a segmented design along the range can be completed. Doppler deblurring must be performed first during the imaging processing of each imaging sub-block, followed immediately by imaging processing. For example... Figure 6 The diagram shows the implementation flowchart of a parallel fast imaging processing method for Ka-band high-resolution wide-swath SAR.

[0092] The following sections will elaborate on the virtual equivalent direct sampling method for FMCW deslant (beat) receiver systems, the large sampling data dimensionality reduction processing technology for spaceborne large slant angle ultra-high resolution SAR imaging, and the ultra-high resolution SAR fast imaging method based on range-direction frequency domain downsampling proposed in this invention.

[0093] Assume the received signal after azimuth segmentation and track curvature compensation is:

[0094]

[0095] in, This indicates the received signal after azimuth segmentation and track curvature compensation. Representing relative time in the fast time interval, with the transmitted pulse as the reference time, t m R represents slow time. t The distance from the target to the antenna is represented by c, where c represents the speed of light, and T represents the distance from the target to the antenna. p f represents the radar transmit pulse width. c γ represents the carrier frequency of the radar transmitted signal, t represents the absolute time in fast time, and γ represents the radar frequency modulation slope.

[0096] Since the bandwidth after deskewing reception depends on the distance difference between the radar's distance to the target at the azimuth sampling position and the reference slant range corresponding to the reference signal, the larger the distance difference, the larger the bandwidth after deskewing reception. Therefore, reducing this distance difference can effectively reduce the bandwidth after deskewing reception, thereby reducing the subsequent sampling rate.

[0097] For each azimuth sampling point, the distance from that sampling point to the center point of the imaging scene is used as the reference distance. This means that the reference slant range is different for each azimuth sampling point. This variable reference slant range processing method reduces the range direction (the direction the antenna points towards the target is called the range direction, and the direction the antenna moves forward is called the azimuth direction) sampling bandwidth corresponding to each azimuth sampling point, thereby reducing the number of range direction sampling points and reducing memory consumption. Different reference slant ranges are used for de-slant reception, that is, variable reference slant range de-slant reception is applied to the signal after overall trajectory curvature compensation.

[0098]

[0099] Among them, R Δi =R t -R ref_i R ref_i For the i-th reference slope distance, This represents the reference signal corresponding to the i-th reference slant distance.

[0100] For fast time in Equation 10 (Using the time at the reference point as the baseline, with the time along the distance direction as fast time and the time along the azimuth direction as slow time), a Fourier transform is performed to obtain:

[0101]

[0102] Where A represents amplitude, f i Represents frequency, S if (f i ,t m The ) refers to the Fourier transform of the previous expression in the fast time domain.

[0103] In the slow time domain, that is, at each t... m Phase compensation is performed on Equation 11, i.e., multiplied by get:

[0104]

[0105] The echo signal expression after direct sampling matched filtering is:

[0106]

[0107] Among them, A IQ Where B is the amplitude value, and R is the coefficient. Δ This refers to the distance difference between the radar's distance to the target at the azimuth sampling position and the reference distance corresponding to the reference signal.

[0108] Therefore, it is necessary to study the conversion relationship from Equation 12 to Equation 13, and then transform the echo signal after deslant reception into a virtual equivalent echo signal after direct sampling matched filtering. Afterwards, the imaging processing algorithm after direct sampling matched filtering can be used entirely for imaging processing. Comparing Equations 12 and 13, Equation 12 first needs to undergo secondary phase compensation to make it equivalent to the phase under a unified reference slant range in Equation 13. Let the compensation factor be... Therefore:

[0109]

[0110] Where c represents the speed of light, f c R represents the carrier frequency of the antenna's transmitted signal. Δ R represents the distance difference between the antenna at the azimuth sampling position and the target, and the reference distance corresponding to the reference signal. Δi =R t -R ref_i R t R represents the distance from the target to the antenna. ref_i This is the i-th reference slope distance.

[0111] Immediately following Equation 14 in f i The domain performs filtering on the corresponding sub-block to be imaged within the entire range cell (this step is related to s(ω,t) after Equation 25). m (After downsampling in the ω domain, this is combined to reduce the number of computation points and thus reduce the computational load. That is, the range gate containing the range imaging patch retains the data, while the other range gates are zeroed out.)

[0112] S if_com3 (f i ,t m ) = S if_com2 (f i ,t m )H filter (f i (15)

[0113] Among them, H filter (f i ) is the filtering function, and then the target in Equation 15 needs to be filtered in the frequency domain f. i Domain flip translation is the translation amount under a unified reference slant range. After flip translation, it becomes:

[0114]

[0115] Among them, R Δ This refers to the distance difference between the radar's distance to the target at the azimuth sampling position and the reference distance corresponding to the reference signal.

[0116] When performing the above translation operation in the digital domain, it is necessary to calculate the translation unit in the digital domain, because:

[0117]

[0118] Where dt is the time-domain sampling interval during deskewing reception, and ω move k is the shift of the digital angular frequency. move Let N be the translation amount of the digital unit, and N be the number of de-skewing sampling points. Therefore:

[0119]

[0120] After the translation is complete, the transformation relationship between Equation 16 and Equation 13 in the digital domain needs to be studied. For Equation 16, the translation amount corresponding to the unified reference slant distance is:

[0121]

[0122] For Equation 13, the translation amount corresponding to the unified reference slant distance is:

[0123]

[0124] dt' is the temporal sampling interval under virtual equivalent direct sampling, k' move_ref This represents the digital shift amount under direct sampling, which is virtually equivalent. To make Equations 13 and 16 equivalent, the following must be satisfied:

[0125] k move_ref =k' move_ref (twenty one)

[0126] Therefore:

[0127]

[0128] From this point on, subsequent imaging processing can be performed according to the imaging algorithm under the time-domain sampling interval dt' of virtual equivalent direct sampling.

[0129] In spaceborne high-frequency band (such as Ka band) ultra-high resolution SAR imaging, such as Figure 7 As shown, to obtain a wide-swath SAR image, it is necessary to perform block imaging using forward-looking, side-looking, and backward-looking sliding spotting modes. For any imaging point, the corresponding effective synthetic aperture length is relatively long. In the case of oblique looking, when performing imaging processing using the imaging algorithm under the above-mentioned virtual equivalent direct sampling time-domain sampling interval dt', the number of range-traveling units is relatively large (e.g., Figure 8 As shown in the figure, the number of data processing matrices increases dramatically in the distance direction.

[0130] To reduce the number of processing units in the distance direction and thus reduce the computational load, such as Figure 8As shown, the reference coordinate system is adjusted accordingly to the equivalent frontal side view under the oblique angle. By projecting the azimuth sampling point onto a straight line perpendicular to the oblique angle, it becomes a virtual synthetic aperture, i.e., the azimuth sampling point S is projected to S'. Under this virtual synthetic aperture, the imaging area can be equivalent to the frontal side view under the new coordinate system, which is equivalent to correcting the distance travel portion (e.g., ...). Figure 10 As shown, the distance migration curve after correction and compression is shown (the span of the entire migration cell is significantly reduced after correction). Then, the frontal side view imaging algorithm is used for imaging processing. After obtaining the imaging result, the imaging result is rotated according to the oblique angle to finally obtain the imaging result in the original coordinate system. This method can significantly reduce the memory consumption.

[0131] In the equivalent frontal side view processing, R in Equations 13 and 16 Δ This will become R' in the equivalent frontal side view case, i.e., the equivalent reference slant range corresponding to the virtual projection sampling point S'. Simultaneously, the virtual sampling step size becomes:

[0132] du'=du·cos(theta_c) (23)

[0133] After coordinate system transformation, the virtual sampling step size and other transformed time-frequency variables are determined based on the coordinates of the sampling points after geometric transformation. Then, imaging can be performed using the imaging algorithm under the original frontal and side views.

[0134] Replace the symbols in Equation 16 (replace f) i Replace with (This step does not involve calculation), resulting in:

[0135]

[0136] Where A represents amplitude, T p γ represents the radar transmit pulse width, and γ represents the antenna frequency modulation slope.

[0137] Next, each imaging sub-block needs to be imaged. When processing the imaging of each imaging sub-block, Doppler deblurring must be performed first, followed by imaging processing.

[0138] make Where, k = ω / c, X c Y c and R c The coordinates of the center reference point of the corresponding imaging sub-block are first applied to Equation 24 in the fast time domain (i.e., Perform a Fourier transform on the domain (ω,t) to obtain s(ω,t). m ),Right now:

[0139]

[0140] in, Indicates to exist Perform a Fourier transform on the domain.

[0141] For ultra-high resolution imaging, to reduce curved trajectory errors, the number of range blocks increases accordingly. If the same number of sampling points is used in the ω-domain, the data volume increases dramatically. Therefore, methods to reduce the data volume need to be investigated. Since the range swath of the corresponding imaging sub-blocks decreases with the increase in range blocks, this prior information can be used to increase the sampling interval in the ω-domain (i.e., the angular frequency domain). Specifically, the sampling interval of s(ω,t) can be increased according to the multiple of the range blocks. m Downsampling is performed in the ω-domain, with the downsampling factor equal to the distance-oriented block factor, thereby reducing s(ω,t) m This reduces the number of sampling points in the ω-domain, thus reducing computational complexity. Then, in the slow time domain (i.e., t...) m Compressing the domain yields:

[0142]

[0143] For Equation 26 in the slow time domain (i.e., t) m Performing a Fourier transform on the domain, we get:

[0144]

[0145] In k u The domain is upsampled by adding zeros, resulting in S. c (ω,k u ) becomes S cd (ω,k u ). For S cd (ω,k u ) in k u Performing an inverse Fourier transform on the domain yields:

[0146]

[0147] Decompressing equation 28 yields:

[0148] s d (ω,t m ) = s cd (ω,t m )s ref (ω,t m (29)

[0149] For equation 29, at t m Doppler frequency shift correction is performed in the domain, i.e.:

[0150] s d1 (ω,t m ) = s d(ω,t m )exp(-j*2*k c *sin(theta_c)*vt m (30)

[0151] theta_c is the oblique angle of this sub-block. When using a variable coordinate system, the oblique angle can be equivalent to 0.

[0152] The imaging process is then performed. Equation 30 is applied in the slow time domain (i.e., t...). m Performing a Fourier transform on the domain, we obtain:

[0153]

[0154] For S in Equation 31 11 (ω,k u ) in k u Sampling is performed on the domain, i.e., downsampling, to obtain S(ω,k). u To ensure that coordinate system one is centered at the reference point, let... Then we have:

[0155]

[0156] make k y (ω,k u )=k u , for S o (ω,k u Perform interpolation (after interpolation F) s Points can be compared with S o (ω,k u (Mainly consistent) we get:

[0157] F s =F[k x (ω,k u ),k y (ω,k u (33)

[0158] For F s Perform a two-dimensional inverse Fourier transform to obtain the imaging results:

[0159]

[0160] The imaged result is then rotated in the coordinate system, i.e.:

[0161] f(x,y)=rot(f(x,y) theta_c (35)

[0162] After imaging the sub-blocks, the next step is to stitch the individual sub-blocks together in the image domain to finally complete the full-width image.

[0163] In summary, this invention proposes a parallel and fast imaging processing method for high-frequency (such as Ka-band) spaceborne high-resolution wide-swath SAR. By establishing the geometric relationship of spaceborne wide-swath high-resolution imaging, designing the instantaneous illumination range of the antenna, and then constructing independent imaging sub-blocks in the azimuth and range directions, parallel and fast imaging is finally completed on the GPU.

[0164] Meanwhile, this invention proposes a virtual equivalent direct sampling method for FMCW deslant (beat) reception systems. By establishing the correlation transformation relationship between the echo signal after direct sampling and the signal after deslant reception, FMCW deslant reception data can be processed based on the imaging processing method of direct sampling.

[0165] Moreover, this invention proposes a large sampling data dimensionality reduction processing technique for spaceborne large oblique angle ultra-high resolution SAR imaging. By combining a variable coordinate system processing method with a variable reference slant range sampling data dimensionality reduction preprocessing method, the amount of sampling data can be significantly reduced.

[0166] In addition, this invention proposes a fast satellite-borne ultra-high resolution SAR imaging method based on range-direction frequency domain downsampling. Under the preprocessing framework of variable reference slant range and variable coordinate system, the method first performs overall pulse compression in the range direction, then divides the image into blocks in the pulse compression domain, and then performs range-direction time domain pre-filtering, and finally performs range-direction frequency domain downsampling, which can significantly reduce the computational load of block imaging processing.

[0167] In high-frequency (e.g., Ka-band) spaceborne sliding spotting high-resolution SAR imaging with a wide azimuth swath, the traditional whole-block imaging processing method results in a huge number of azimuth points, leading to a sharp increase in computational load. This invention proposes a parallel fast imaging processing method for high-frequency (e.g., Ka-band) spaceborne high-resolution wide-swath SAR, which can significantly improve computational efficiency.

[0168] Traditional pulsed SAR typically employs direct sampling reception. However, when transitioning to ultra-high resolution (e.g., 0.05m) sliding spotting mode SAR imaging, the amount of sampled data increases dramatically. This invention proposes a virtual equivalent direct sampling method for FMCW deslant (beat) reception systems, which can significantly reduce the amount of sampled data. In spaceborne ultra-high resolution deslant reception SAR imaging, due to the high resolution and long synthetic aperture length, using traditional unified reference slant range processing methods results in a sharp increase in range-direction sampling bandwidth, leading to a sharp increase in the sampling rate. Furthermore, under large slant-view conditions, the data matrix resulting from ultra-high resolution range migration increases dramatically in the range direction. This invention's proposed dimensionality reduction technology for large sampled data in spaceborne large slant-view ultra-high resolution SAR imaging can significantly reduce the amount of sampled data. In addition, the proposed ultra-high resolution SAR fast imaging method based on range-direction frequency domain downsampling can significantly reduce the computational load during block imaging processing.

Claims

1. A spaceborne sliding spotting mode SAR imaging method based on ultra-high resolution FMCW system, characterized in that, Includes the following steps: Receive the reflected signals from sub-blocks of the region to be imaged; The fast time portion of the reflected signal is subjected to Fourier transform, and then the slow time portion is phase compensated to obtain signal S. if_com1 (f i ,t m ); where f i Represents frequency, t m Indicates slow time; S if_com1 (f i ,t m The echo signal S is converted into a virtual equivalent direct sampling matched filter. if_com2 (f i ,t m ); For S if_com2 (f i ,t m Perform range pre-filtering to obtain S if_com3 (f i ,t m ); S if_com3 (f i ,t m The target in the frequency domain f i The domain flip translation is the translation amount under a unified reference slant range, resulting in S. if_com4 (f i ,t m ); S if_com4 (f i ,t m (This is obtained by performing fast time domain sign and scaling substitution) in, Indicates a fast time; Will The distance is obtained by frequency conversion in the time domain to obtain s(ω,t) m ), and for s(ω,t m Downsampling is performed in the ω-domain; where ω represents the angular frequency domain; The downsampled signal is imaged to obtain a sub-block image; By fusing several sub-block images, a SAR image of the region to be imaged is obtained.

2. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 1, characterized in that, Before receiving the reflected signal from a sub-block of the region to be imaged, the following steps are also included: Calculate the distance deviation between the circular orbit and the approximately equivalent straight orbit of the SAR antenna at different apertures and different imaging points; When the distance deviation within the imaging area of ​​the sub-block is not greater than a threshold, the imaging area of ​​the sub-block is determined.

3. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 2, characterized in that, After receiving the reflected signal from the sub-block of the region to be imaged, the process also includes: Overall orbital curvature compensation is performed on the imaging area of ​​the sub-block.

4. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 3, characterized in that, After performing overall orbital curvature compensation on the imaging area of ​​the sub-block, the process also includes: The signal after overall track curvature compensation is received by variable reference slant range de-slant reception.

5. A spaceborne sliding spotting mode SAR imaging method for ultra-high resolution FMCW system as described in any one of claims 2-4, characterized in that, For S if_com2 (f i ,t m Perform range pre-filtering to obtain S if_com3 (f i ,t m )include: S if_com3 (f i ,t m )=S if_com2 (f i ,t m )H filter (f i ), Among them, H filter (f i ) is a filtering function, which retains data for the distance gate where the sub-block is located, and sets the other distance gates to zero.

6. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 5, characterized in that, Will The distance in the time domain and frequency conversion domain includes: in, Indicates to exist Perform a Fourier transform on the domain.

7. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 4, characterized in that, The variable reference slant range de-slant reception of the signal after overall track curvature compensation includes: in, This indicates the signal after overall track curvature compensation. This represents the reference signal corresponding to the i-th reference slant distance.

8. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 6, characterized in that, S if_com1 (f i ,t m The echo signal S is converted into a virtual equivalent direct sampling matched filter. if_com2 (f i ,t m )include: Where c represents the speed of light, f c R represents the carrier frequency of the antenna's transmitted signal. Δ R represents the distance difference between the antenna at the azimuth sampling position and the target, and the reference distance corresponding to the reference signal. Δi =R t -R ref_i R t R represents the distance from the target to the antenna. ref_i This is the i-th reference slope distance.

9. The ultra-high resolution FMCW system spaceborne sliding spotting mode SAR imaging method as described in claim 8, characterized in that, S if_com4 (f i ,t m Specifically: Where A represents amplitude, T p γ represents the radar transmit pulse width, and γ represents the antenna frequency modulation slope.

Images