A gcbp imaging method and system of a non-normal side array multi-channel sar of an ultrahigh-speed diving platform

Abstract

The application discloses a GCBP imaging method and system of a non-normal side array multichannel SAR of an ultrahigh-speed diving platform, and the method comprises the following steps: constructing a non-normal side array multichannel SAR geometric model; performing pulse compression and sub-aperture division on echo signal data; constructing a multichannel steering matrix and calculating a weight vector of reconstructed sub-aperture echo signal; constructing a sidelobe suppression window function; traversing all grid points in an imaging area, substituting the weight vector and the sidelobe suppression window function into integral operation of BP imaging, performing BP imaging processing on echo signal in the sub-aperture, and sequentially performing two-stage spectrum compression, azimuth up-sampling and spectrum decompression processing on the sub-aperture multichannel BP image after coherent accumulation; and traversing all sub-apertures, coherently fusing all sub-aperture BP images, and obtaining a non-lattice lobe high-resolution SAR image; and the multichannel steering matrix and the sidelobe suppression window function are fused in the sub-aperture BP integral operator, so that ghosting can be directly eliminated in the BP imaging process, and the anti-fuzzy performance of the SAR image is improved.

Description

Technical Field

[0001] This invention belongs to the field of radar imaging technology, specifically relating to a GCBP imaging method and system for non-frontal array multi-channel SAR suitable for hypersonic diving platforms. Background Technology

[0002] Synthetic Aperture Radar (SAR) possesses all-weather, all-day Earth observation capabilities, making it highly valuable in military reconnaissance, terrain mapping, and target surveillance. With the development of hypersonic vehicle platforms, SAR is increasingly being deployed on hypersonic dive platforms for Earth observation missions. These platforms, characterized by high speed and complex trajectories, enable rapid, wide-area imaging, significantly enhancing combat and reconnaissance capabilities.

[0003] Hypersonic dive platforms equipped with SAR can achieve wide-swath Earth observation, but their extremely wide Doppler bandwidth easily leads to two-dimensional ambiguity in both range and azimuth directions during wide-strip imaging. According to the principle of minimum antenna area, increasing the equivalent antenna area is key to alleviating this contradiction. However, due to limitations in platform shape and surface size, it is difficult to install a single large-area planar antenna. To fully utilize the space on the aircraft surface, a multi-channel SAR system with single-transmitter and multiple-receiver capabilities can be employed. In such a system, the system operates with a non-central array, failing to meet the requirements of a Displaced Phase Center Antenna (DCPA). Consequently, signal sampling in space often exhibits periodic and non-uniform characteristics, and the spatiotemporal spectrum distortion leads to grating lobes and target ghosting during direct multi-channel coherent accumulation.

[0004] Existing multi-channel SAR frequency domain imaging methods mostly rely on azimuth spectrum reconstruction, converting multi-channel data into single-channel data with an equivalent high pulse repetition frequency, followed by frequency domain imaging algorithms for image processing. However, this algorithm is complex, unsuitable for real-time operation, and difficult to adapt to mobile platforms. While the back projection (BP) algorithm can adapt to complex trajectories, it suffers from spatial non-uniform sampling and severe grating lobes under non-frontal side-looking observations such as hypersonic dives due to the failure to meet the DPCA condition. To address this, Zhou et al. (Zhou L, Zhang X, Wang Y, et al. Unambiguous reconstruction for multichannel nonuniform sampling SAR signal based on image fusion[J]. IEEE Access, 2020,8: 71558-71571.) proposed an unambiguous reconstruction method for multi-channel nonuniform sampling SAR based on image fusion, utilizing the generalized sampling theorem and constructing weighted sub-images to obtain unambiguous SAR images. However, this method is limited by computational complexity, and the interpolation kernel function needs to be approximated, which limits the imaging accuracy. Furthermore, it is mainly designed for frontal and side array geometry and is difficult to adapt to non-frontal and side array SAR systems. It cannot solve the complex imaging problems caused by spatiotemporal spectral distortion in non-frontal and side array multi-channel SAR. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the present invention aims to provide a GCBP imaging method and system for non-frontal array multi-channel SAR of hypersonic diving platforms. By incorporating a multi-channel steering matrix and a sidelobe suppression window function into the sub-aperture BP integral operator, ghosting can be eliminated directly during BP imaging without additional spectral reconstruction operations.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A GCBP imaging method for a non-frontal array multi-channel SAR of a hypersonic diving platform includes the following steps: Step 1: Construct a non-frontal array multi-channel SAR geometric model based on the imaging scene information; Step 2: Perform pulse compression and sub-aperture division on the echo signal data received by the radar to obtain echo signals within multiple sub-apertures; Step 3: For each grid point in the imaging region, construct a multi-channel steering matrix and calculate the weight vector of the reconstructed sub-aperture echo signal; Step 4: Construct a sidelobe suppression window function based on the effective synthetic aperture length; Step 5: Traverse all grid points in the imaging region, substitute the weight vector and sidelobe suppression window function into the BP imaging integral calculation, and perform BP imaging processing on the echo signal within the sub-aperture to obtain... Sub-aperture images of each channel, then Coherently accumulate the sub-aperture BP images of each channel to obtain a low-resolution sub-aperture multi-channel BP image without raster lobes. Step 6: Perform two-stage spectral compression, azimuth upsampling, and despectral compression processing on the sub-aperture multi-channel BP image to obtain a lobe-free high-resolution sub-aperture BP image; Step 7: Traverse all sub-apertures and repeat steps 5 to 6 to coherently fuse all sub-aperture BP images to obtain a lobe-free, high-resolution SAR image.

[0007] Constructing a non-frontal array multi-channel SAR geometric model includes: taking the ground projection position of the radar array transmission channel at the radar aperture center at the moment of flight as the origin, and the direction of flight as the origin. Establish a Cartesian coordinate system on the ground. Set the initial position of the high-speed mobile platform. , Platform height, velocity vector acceleration vector Furthermore, the direction of acceleration is the same as the direction of velocity, and the platform's dive angle Arbitrary scattering point in the imaging scene The coordinates are Establish a platform motion model; The signal from the radar array's transmission channel is equivalent to... The first equivalent phase center, the... The coordinates of the equivalent phase centers relative to the channel array center are: Define the channel direction tilt angle as The angle between the platform's dive trajectory direction and the channel arrangement direction is... Defined as: (1) when At this time, the conditions for non-positive side array observation are constituted; Arbitrary scattering point in the imaging scene To the The instantaneous slant distance of the phase center of each channel is expressed as: (2) in, For a slow time along the azimuth direction, the 4th order Maclaurin expansion of the slant range is as follows: (3) in, From the center time to the scattering point The slope distance; (4) Definition of the first The delay caused by the distribution distance between each channel and the reference channel for: (5) in, For the first The first channel and the first The projection of the spacing between the channels in the flight direction. This refers to the platform's flight speed.

[0008] Step 2, the process of pulse compression of the echo signal data received by the radar, includes: The radar baseband echo signal is obtained by down-converting the received radar echo signal data, and its expression is as follows: (6) in, For the fastest time along the distance direction, The slow time along the azimuth direction, The pulse width. To adjust the frequency, At the speed of light, For wavelength, For scattering points The backscattering coefficient; Based on the radar baseband echo signal, a range-direction system matching function is constructed, the expression of which is: (7) Based on the range-direction system matching function, the radar baseband echo signal is pulse-compressed. After pulse compression, the echo signal received by the reference channel phase center is: (8) in, For bandwidth.

[0009] The calculation process for the weight vector of the reconstructed sub-aperture echo signal is as follows: Step 3.1: Calculate the Doppler center frequency of the platform. The calculation formula is as follows: (9) in, Let be the coefficients of the first-order term in the Taylor expansion of the slant distance history. Wavelength; Step 3.2: Based on the maximum Doppler frequency corresponding to the entire imaging grid and minimum Doppler frequency The Doppler bandwidth of the imaging scene is defined by the following expression: (10) Step 3.3: Based on the Doppler bandwidth of the imaging scene, fold the Doppler center frequency and search for all integers that satisfy the folding condition. This ensures that the center frequency of the folded Doppler falls within the Doppler bandwidth of the imaging scene. The folding condition is as follows: (11) in, The pulse repetition frequency; Define fuzzy weight vector All integers that satisfy the folding condition To form a vector: (12) in, and These are the smallest and largest fuzzy integers found, respectively. For imaging grid points The center of the fuzzy weight vector; Step 3.4: Based on the fuzzy weight vector center Constructing imaging grid points Multi-channel steering matrix at the location ; (13) Among them, matrix elements Defined as: (14) In non-frontal and diving situations, the phase term Based on the Doppler center frequency, the corrected expression is as follows: (15) in, For the first The delay caused by the distribution distance between each channel and the reference channel The angle between the platform's dive trajectory direction and the channel arrangement direction. It is an oblique perspective; respectively along The velocity vector in the direction; Step 3.5: Calculate the weight vector of the reconstructed sub-aperture echo signal, using the following formula: (16) in, It is a multi-channel steering matrix.

[0010] The expression for the sidelobe suppression window function is as follows: (17) in, For imaging grid points The location center time, For the length of the window, , , All values ​​are window weights.

[0011] The integration process of the BP imaging can be expressed as: (18) in, After pulse compression, the first Sub-aperture echo signals of each channel, For imaging grid points The corresponding two-way echo delay, This is the phase compensation term before BP integration. To reconstruct the weight vector of the sub-aperture echo signal, This is a sidelobe suppression window function. for At this moment Each channel to the imaging grid point Instantaneous slant distance, This is the slow time along the azimuth direction.

[0012] The sub-aperture multi-channel BP image is subjected to two-stage spectral compression, azimuth upsampling, and despectral compression processing, including: Sub-aperture multichannel BP image with first-level spectral compression function Multiply to obtain the BP image after spectral support region alignment; Perform a range-direction FFT on the BP image after spectral support region alignment, and it becomes... Domain, and the second-order spectral compression function Multiplying them yields the BP image after spectral tilt correction; Perform azimuth FFT on the BP image after spectral tilt correction to obtain... In the wavenumber domain, wavenumber compression is completed, resulting in a spectrally compressed BP image. The BP image after spectral compression is zero-padding at both ends of the two-dimensional beam spectrum to complete azimuth upsampling; The upsampled BP image was subjected to two-stage despectral compression to obtain a high-resolution sub-aperture BP image without grating lobes.

[0013] The expression for the two-stage spectral compression function is as follows: (20) in, For the range wave number, carrier frequency The corresponding wave number, At the speed of light, This serves as the virtual beam rotation center.

[0014] This invention also provides a GCBP imaging system for a non-frontal array multi-channel SAR of a hypersonic diving platform, comprising: Model building module: Constructs a non-orthogonal side array multi-channel SAR geometric model based on imaging scene information; Data segmentation module: Performs pulse compression and sub-aperture segmentation on the echo signal data received by the radar to obtain echo signals within multiple sub-apertures; Weight vector calculation module: For each grid point in the imaging area, construct a multi-channel steering matrix and calculate the weight vector of the reconstructed sub-aperture echo signal; Window function construction module: Construct a sidelobe suppression window function based on the effective synthetic aperture length; BP Imaging Module: Traverses all grid points in the imaging region, substitutes the weight vector and sidelobe suppression window function into the integral calculation of BP imaging, and performs BP imaging processing on the echo signal within the sub-aperture to obtain... Sub-aperture images of each channel, then Coherently accumulate the sub-aperture BP images of each channel to obtain a low-resolution sub-aperture multi-channel BP image without raster lobes. Despectral compression module: Performs two-stage spectral compression, azimuth upsampling and despectral compression processing on the sub-aperture multi-channel BP image in sequence to obtain a lobe-free high-resolution sub-aperture BP image; Image fusion module: Iterates through all sub-apertures and coherently fuses all sub-aperture BP images to obtain a lobe-free, high-resolution SAR image.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Strong Anti-Blur Targeting: This invention is based on the imaging perspective of the ground-based Cartesian coordinate system, focusing on the spatial non-uniform sampling characteristics and Doppler spectral aliasing phenomenon under non-frontal array observations of ultra-high-speed diving platforms. By constructing a multi-channel steering matrix to suppress grating lobes and synergistically fusing it with a sidelobe suppression window function, it can accurately capture the interference of real targets and grating lobes (ghosting) formed by non-uniform sampling on multi-channel BP imaging. It achieves accurate identification and efficient suppression of Doppler blur components during the sub-aperture BP imaging integration process. Compared with traditional direct multi-channel BP imaging methods, this invention improves the anti-blurring performance of SAR images, significantly reduces the impact of false targets on BP imaging, and significantly enhances the imaging purity and continuous stable operation capability of multi-channel SAR in ultra-high-speed maneuvers and complex non-frontal array scenarios.

[0016] 2. High Efficiency and Real-Time Performance: This invention overcomes the bottleneck of the enormous computational burden of traditional time-domain BP algorithms. It employs an improved GCBP algorithm based on multi-channel SAR, decomposing the high-dimensional integration operation of the entire aperture into low-dimensional sub-aperture processing through a sub-aperture partitioning strategy, significantly reducing the algorithm's time complexity. By introducing the multi-channel steering matrix into the direct integration operation of BP imaging, this invention eliminates image lobes caused by non-uniform spatial sampling without additional operations, meeting the real-time image processing requirements of ultra-high-speed platforms. The real-time performance and high efficiency of this invention enable its widespread application in fields requiring rapid response, such as military reconnaissance and disaster monitoring.

[0017] In summary, compared with existing technologies, this invention, by incorporating a multi-channel steering matrix and a sidelobe suppression window function into the sub-aperture BP integral operator, eliminates ghosting directly during BP imaging without additional spectral reconstruction operations. This provides a novel imaging method for SAR systems on high-speed diving platforms, offering advantages such as high precision, high resolution, and high efficiency, while maintaining excellent imaging performance even in complex interference environments. Its broad engineering applicability gives this technology significant commercial value and allows for effective applications in various fields, including military, meteorology, and disaster monitoring. Attached Figure Description

[0018] Figure 1 This is a flowchart of the overall algorithm of the present invention.

[0019] Figure 2 This is the non-frontal array multi-channel SAR geometric model of the ultra-high speed diving platform of the present invention.

[0020] Figure 3 This is a flowchart of the GCBP algorithm based on multi-channel SAR of the present invention.

[0021] Figure 4 This is a multi-channel single-target BP imaging result image of the present invention; wherein, Figure 4 (a) in the image shows the BP imaging results of the traditional direct multi-channel BP imaging algorithm. Figure 4 (b) in the figure is an azimuth profile of the equidistant ring where the target is located in the traditional direct multi-channel BP imaging algorithm. Figure 4 (c) in the figure represents the BP imaging result of the BP imaging algorithm of the present invention. Figure 4 (d) in the figure is the azimuth profile of the equidistant circular ring where the target of the BP imaging algorithm of the present invention is located.

[0022] Figure 5 This is the large-scene multi-target imaging result of the present invention; wherein, Figure 5 (a) in the image shows the BP imaging results of the traditional direct multi-channel BP imaging algorithm. Figure 5(b) in the figure represents the BP imaging result of the BP imaging algorithm of the present invention.

[0023] Figure 6 The image shows the point target imaging result of the BP imaging algorithm of this invention; wherein, Figure 6 (a) in the figure is the contour plot of the point target without a window function. Figure 6 (b) in the diagram is the azimuth profile of the point target without a window function. Figure 6 (c) in the figure is the contour plot of the point target after applying the window function. Figure 6 (d) in the figure is the azimuth profile of the point target after applying the window function. Detailed Implementation

[0024] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0025] like Figure 1 As shown, a GCBP imaging method for a non-frontal array multi-channel SAR of a hypersonic diving platform specifically includes the following steps: Step 1: Based on the imaging scene information, construct a non-frontal array multi-channel SAR geometric model of the hypersonic diving platform to be suitable for airborne hypersonic maneuvering platform SAR observation; Step 1.1: Establish the ground rectangular coordinate system (GC) and the platform motion model; The high-speed maneuvering platform (hereinafter referred to as "the platform") is in a high-speed dive motion. Assuming the platform's attitude remains unchanged for a short period of time, with the ground projection position of the radar array transmission channel (hereinafter referred to as "the channel") at the center of the radar aperture as the origin, and the flight direction as... Establish a Cartesian coordinate system on the ground. The initial position of the platform is represented as ,in The platform height is given by the velocity vector. The acceleration vector is Furthermore, the direction of acceleration is the same as the direction of velocity, and the platform's dive angle is defined as... Arbitrary scattering point in the imaging scene The coordinates are Its geometric diagram is as follows Figure 2 As shown; Step 1.2: Construct a non-frontal multi-channel SAR geometric model; This invention uses a "one-transmit, multiple-receive" model, utilizing the Effective Phase Center (EPC) principle to equate the signal of the radar array's transmitting channel to... The first equivalent phase center, the... The coordinates of the equivalent phase centers relative to the channel array center are: Define the channel direction tilt angle as The angle between the platform's dive trajectory direction and the channel arrangement direction is... Defined as: (1) when At this time, the conditions for non-positive side array observation are constituted; Arbitrary scattering point in the imaging scene To the The instantaneous slant distance of the phase center of each channel can be expressed as: (2) in, For a slow time along the azimuth direction, the 4th order Maclaurin expansion of the slant range is as follows: (3) in, From the center time to the scattering point The slope distance; (4) No. The echo signal of the first channel is the reference channel signal after a certain time delay, and the first channel is defined as follows: The delay caused by the distribution distance between each channel and the reference channel for: (5) in, For the first The first channel and the first The projection of the spacing between the channels in the flight direction. This refers to the platform's flight speed.

[0026] Step 2: Echo data pulse compression and sub-aperture segmentation; Assuming the radar transmits a linear frequency modulated (LFM) signal, and the received raw echo signal is full-aperture multi-channel echo signal data, the radar baseband echo signal is obtained by down-converting the received echo signal data, and its expression is as follows: (6) in, For the fastest time along the distance direction, The pulse width. To adjust the frequency, At the speed of light, For wavelength, For scattering points The backscattering coefficient.

[0027] Based on the radar baseband echo signal, a range-direction system matching function is constructed, the expression of which is: (7) Based on the range-direction system matching function, the radar baseband echo signal is pulse-compressed. After pulse compression, the echo signal received by the reference channel phase center is: (8) in, For bandwidth.

[0028] To reduce the computational complexity of subsequent BP imaging, the pulse-compressed echo signal was then processed into sub-aperture blocks in the azimuth direction to obtain echo data within multiple sub-apertures, thereby improving imaging efficiency.

[0029] Step 3: Based on the non-frontal multi-channel SAR geometric model, set the ground resolution and divide the imaging area into several grids; for each grid point in the imaging area, construct a multi-channel steering matrix that can separate the energy of the real target from the energy of the grating lobe (ghost image) according to the spatial location of the imaging scene and the geometric relationship of the multi-channel radar, and calculate the weight vector of the reconstructed sub-aperture echo signal. Step 3.1: Based on the relative motion between the imaging grid points and the radar, calculate the Doppler center frequency of the platform. The calculation formula is as follows: (9) in, For the coefficients of the first-order term in the Taylor expansion of the slant distance history; Step 3.2: Based on the maximum Doppler frequency corresponding to the entire imaging grid and minimum Doppler frequency The Doppler bandwidth of the imaging scene is defined by the following expression: (10) Step 3.3: Based on the Doppler bandwidth of the imaging scene, fold the Doppler center frequency and search for all integers that satisfy the folding condition. This ensures that the center frequency of the folded Doppler falls within the Doppler bandwidth of the imaging scene. The folding condition is as follows: (11) in, is the pulse repetition frequency.

[0030] Define fuzzy weight vector All integers that satisfy the folding condition To form a vector: (12) in, and These are the smallest and largest fuzzy integers found, respectively. For imaging grid points The center of the fuzzy weight vector; Step 3.4: Based on the fuzzy weight vector center and channel delay Construct the imaging grid points Multi-channel steering matrix at the location ; Multi-channel steering matrix The dimension is ,in For the number of channels, The number of fuzzy weight vectors, multi-channel steering matrix elements Indicates the first The first channel is for the first Phase influence of a fuzzy weight vector Let be the phase response function, representing the amplitude variation of noise frequencies caused by non-uniform spatial sampling of multi-channel signals. For a non-frontal array diving platform, the phase response function is... It must include a strabismus correction term, and its specific construction form is as follows: (13) Among them, matrix elements Defined as: (14) Under ideal frontal side view, the phase term In non-frontal and diving situations, the phase term It needs to be corrected according to the Doppler center frequency to compensate for the [various effects]. The corrected expression for the additional Doppler shift caused by the dive angle is as follows: (15) in, It is an oblique perspective; Step 3.5: Based on the multi-channel steering matrix Calculate the weight vector of the reconstructed sub-aperture echo signal; To recover the true target energy from the mixed signal received from multiple channels and to suppress the grid lobe (ghosting) energy caused by grid point blurring to the greatest extent, a weight vector based on digital beamforming theory is used to reconstruct the sub-aperture echo signal. It can be represented as: (16) This weight vector will be used directly in the BP integration operation in step 5 to achieve simultaneous imaging and ghosting removal. After extracting the main lobe energy and grating lobe energy using a signal reconstruction algorithm, the fuzzy weight vector center at different grid points is used... The matrix index corresponding to the main lobe energy is obtained, and then the unblurred main lobe energy is extracted. After reconstructing for each grid point, the main lobe energy of the entire scene can be obtained, thus obtaining the unblurred BP image.

[0031] Step 4: Construct the sidelobe suppression window function; Since the radar scattering coefficients of targets in the imaging scene vary greatly, in order to prevent the sidelobes of strong scattering point targets from drowning out the signals of weak targets, it is necessary to construct a sidelobe suppression window function to suppress the sidelobes of point targets.

[0032] For radar array transmit channel beam pointing patterns, the synthetic apertures of different target points are inconsistent, making uniform windowing impossible. However, given the inertial navigation information, the target's beam illumination range and radar trajectory are used as prior conditions. The times when each imaging grid point enters and leaves the radar beam main lobe are calculated, thereby determining the effective synthetic aperture length corresponding to that point. (Expressed in pulse count) and the corresponding azimuth center time.

[0033] Based on the determined effective synthetic aperture length For each imaging grid point The windowing weights corresponding to the current azimuth and time are generated respectively. This invention adopts an adaptive Hamming window form, and the expression of the sidelobe suppression window function is as follows: (17) in, For imaging grid points The location center time, For the length of the window, , , These are all window weights; generally speaking, All are between [0.4, 0.9].

[0034] Step 5: Traverse all grid points in the imaging region and apply the weight vector from Step 3. and the sidelobe suppression window function constructed in step 4 Substituting into the integral calculation of BP imaging, the echo signal within a certain sub-aperture in step 2 is processed using BP imaging to obtain... Sub-aperture images of each channel, then Coherent accumulation of sub-aperture BP images of each channel yields a low-resolution sub-aperture multi-channel BP image without raster lobe (ghosting). Using the direct integration imaging framework, the integration process of BP imaging can be expressed as: (18) in, After pulse compression, the first Sub-aperture echo signals of each channel, For imaging grid points The corresponding two-way echo delay, This is the phase compensation term before BP integration. For the first Each channel to the imaging grid point The instantaneous slant distance at a given moment.

[0035] Due to the sidelobe suppression window function during integration. The gain of the main lobe energy of the true target energy was maximized, while the energy of the grating lobes satisfying the folding condition was minimized. Therefore, the resulting sub-aperture BP image is blur-free (ghost-free). Due to the limitation of the sub-aperture length, the image obtained at this time is a low-resolution image, which needs to be improved by subsequent steps.

[0036] Step 6: In the ground-based Cartesian coordinate system constructed in Step 1, the sub-aperture multi-channel BP image is subjected to two-stage spectral compression, azimuth upsampling, and despectral compression processing to obtain a lobe-free, high-resolution sub-aperture BP image, reducing computation and improving algorithm efficiency. Figure 3 As shown; Step 6.1: Determine the virtual beam rotation center ; For multi-channel beam pointing SAR imaging, the radar and beam pointing are constantly moving, and the position of the beam-illuminated grid is constantly changing; in order to accurately perform spectral compression, it is necessary to determine the virtual beam rotation center. Based on the non-frontal multi-channel SAR geometric model, the beam illumination line of channel 1 at the initial moment and the beam illumination line of channel M at the end of the aperture are extended. The intersection point of the extended beam illumination line and channel M is the virtual beam rotation center. For SAR spotting mode, the beam remains completely illuminating the target imaging area during long-term imaging, and the radar positions of its virtual beam rotation center and sub-aperture center are basically coincident; Beam decomposition can be used to obtain a grid in the imaging scene. The corresponding spectral support region center is expressed as follows: (19) in, For a certain grid in the imaging scene Beam size in the horizontal direction of a Cartesian coordinate system For a certain grid in the imaging scene Beam size along the vertical axis in a Cartesian coordinate system carrier frequency The corresponding wave number, The speed of light; Step 6.2: Construct an improved two-level wavenumber spectrum compression function; Based on the spectral support region determined in step 6.1, and according to the virtual beam rotation center... A two-stage spectral compression function in multi-channel mode is constructed, and its expression is as follows: (20) in, Range wavenumber; Step 6.3: Compare the sub-aperture multi-channel BP image with the first-level spectral compression function. Multiply to obtain the BP image after spectral support region alignment; Step 6.4: Perform a distance-direction FFT on the BP image from Step 6.3, resulting in... Domain, and the second-order spectral compression function Multiplying them yields the BP image after spectral tilt correction; Step 6.5: Perform an azimuth FFT on the BP image from Step 6.4, resulting in... In the wavenumber domain, wavenumber compression is completed, resulting in a spectrally compressed BP image. Step 6.6: Perform zero-padding at both ends of the azimuth direction of the BP image after spectral compression to complete azimuth upsampling; Step 6.7: Perform two-stage despectral compression on the upsampled BP image to obtain a lobe-free, high-resolution sub-aperture BP image; The despectral compression process is as follows: Steps 6.5 to 6.3 are executed in reverse order, performing an azimuth-directed IFFT on the upsampled BP image and multiplying it by the conjugate of the second-level spectral compression function. Then perform a distance-oriented IFFT, multiplying by the conjugate of the first-level spectral compression function. This allows for the recovery of sub-aperture BP images with correct phase and improved resolution. Step 7: Traverse all sub-apertures and repeat steps 5 to 6 to coherently fuse all sub-aperture BP images to obtain a lobe-free, high-resolution SAR image.

[0037] The beneficial effects of the present invention will be further illustrated below with reference to specific embodiments.

[0038] Verification of the effectiveness of the reconstruction weight vector based on point targets: This experiment aims to verify the effectiveness of the weight vector (steering matrix) of the reconstructed sub-aperture echo signal calculated in steps 3 and 5 in eliminating ghosting in multi-channel non-uniform sampling. Specific parameters are shown in the table below: A point target was set 3km to the left of the center of the imaging scene for simulation verification. Both traditional direct multi-channel BP imaging and the weighted BP imaging of this invention were used for processing, and the results are compared as follows: Figure 4 As shown. By Figure 4 As shown in (a) and (b), due to the multi-channel SAR system not satisfying the DPCA condition and the existence of non-positive side array effects, spatial sampling is non-uniform, resulting in ambiguity. Obvious first and second ghost images appear on the right side of the real target, such as... Figure 4 As shown in the red box and annotation in (a) of the document. (By...) Figure 4 As shown in the cross-sectional diagram (b) in the figure, the ghost energy peak is extremely high, which seriously interferes with the identification of the real target and is very likely to cause false alarms.

[0039] Depend on Figure 4 As can be seen from (c) and (d) in the figure, the BP imaging algorithm of the present invention introduces an optimal weight vector in the integration process. Figure 4 The ghostly figure that originally existed in (c) has been completely eliminated. From Figure 4 As shown in the cross-sectional diagram (d), the energy of the false target is suppressed below the noise floor, and only the sharp main lobe of the real target is retained, which verifies the significant effectiveness of the weight vector of the reconstructed sub-aperture echo signal in resisting Doppler blurring.

[0040] Verification of the effectiveness of the large-scene multi-target imaging algorithm: This experiment verifies the comprehensive performance of the improved ground-based Cartesian coordinate system BP algorithm proposed in this invention when handling multiple targets and large scenes, and verifies the effectiveness of the sidelobe suppression window function in step 4 and the improved two-stage spectral compression function in step 6. The experiment is set up in a complex scene with multiple scattering points. This experiment simulates a radar operating in the Ka band, mounted on a near-space hypersonic vehicle. The platform is in a high-maneuver, dive-like state, and the beam pointing exhibits a large squint. Specific radar system parameters, platform motion parameters, and geometric configuration parameters are shown in the table below: Using the above parameters, the Doppler width of the entire imaging scene Approximately 67kHz. If a traditional BP imaging architecture is used, the repetition frequency of the reconstructed pulse needs to be met. Greater than the Doppler width Considering the spatial degree of freedom limitations, the system requires 6 channels. However, the direct BP imaging framework proposed in this invention only requires 3 channels. The imaging results are as follows: Figure 5 As shown, the imaging result of the target at the center of the imaging scene is as follows. Figure 6 As shown.

[0041] like Figure 5As shown in the red dashed box in (a) of the image, multiple strong scattering points in the imaging scene are accompanied by obvious pairs of ghosting shadows. These ghosting shadows overlap and interfere with each other in multi-target scenes, resulting in a decrease in the image signal-to-noise ratio and easily leading to misjudgment of target positions.

[0042] Depend on Figure 5 As shown in (b), ghosting of all targets in the imaging scene was effectively eliminated, and the image contrast was significantly improved. At the same time, after spectral compression, azimuth upsampling, despectral compression and coherent fusion, all point targets still maintained the correct position coordinates and good focusing effect, without geometric distortion or defocusing caused by large squint or high-speed dive.

[0043] Depend on Figure 6 As shown in (a) and (b), when the adaptive Hamming window function is not added, the target azimuth direction exhibits typical characteristics. The function exhibits a high first sidelobe (approximately -13.02 dB) and subsequent sidelobes. These high-intensity sidelobes diffuse outwards, easily masking nearby weakly scattering targets.

[0044] Depend on Figure 6 As shown in (c) and (d), this invention, by incorporating an adaptive Hamming window function, although slightly widening the main lobe, significantly suppresses the sidelobe level with a large amplitude reduction. Measurements show that the peak-to-sidelobe ratio (PSLR) after adding the window function is better than -25dB, and the integral-to-sidelobe ratio (ISLR) is better than -23dB. This indicates that the adaptive Hamming window function can effectively converge the target energy, significantly improve image contrast and dynamic range, and is beneficial for detecting weak targets against strong clutter backgrounds.

[0045] Experimental results show that the algorithm proposed in this invention can effectively overcome the problem of Doppler spectrum aliasing under large oblique viewing of ultra-high speed diving platforms. It can significantly suppress grating lobe interference caused by non-uniform sampling in the sub-aperture imaging stage. At the same time, the adaptive Hamming window function effectively reduces side lobes. While ensuring high-resolution focusing of point targets, it completely eliminates false targets, verifying the effectiveness, robustness and engineering applicability of the method under extreme geometric conditions.

[0046] To address the technical challenges of Doppler blurring, spectral distortion, and conventional imaging algorithm failure in non-frontal array multi-channel SAR operating modes on hypersonic dive platforms, this invention proposes a ground Cartesian coordinate system backward projection imaging method suitable for hypersonic dive platforms. This algorithm constructs a blur-suppressing multi-channel steering matrix and an adaptive azimuth window function, directly integrating them into the sub-aperture BP integral operator. Combined with an improved wavenumber spectrum compression technique based on a virtual beam rotation center, it achieves precise suppression of Doppler blur components and efficient focusing of large squint echoes. Simulation results further verify the reliability of the proposed algorithm, providing robust and efficient technical support for real-time high-precision SAR imaging in hypersonic maneuvers and complex geometric environments.

Claims

1. A GCBP imaging method for a non-frontal array multi-channel SAR of a hypersonic diving platform, characterized in that, Includes the following steps: Step 1: Construct a non-frontal array multi-channel SAR geometric model based on the imaging scene information; Step 2: Perform pulse compression and sub-aperture division on the echo signal data received by the radar to obtain echo signals within multiple sub-apertures; Step 3: For each grid point in the imaging region, construct a multi-channel steering matrix and calculate the weight vector of the reconstructed sub-aperture echo signal; Step 4: Construct a sidelobe suppression window function based on the effective synthetic aperture length; Step 5: Traverse all grid points in the imaging region, substitute the weight vector and sidelobe suppression window function into the BP imaging integral calculation, and perform BP imaging processing on the echo signal within the sub-aperture to obtain... Sub-aperture images of each channel, then Coherently accumulate the sub-aperture BP images of each channel to obtain a low-resolution sub-aperture multi-channel BP image without raster lobes. Step 6: Perform two-stage spectral compression, azimuth upsampling, and despectral compression processing on the sub-aperture multi-channel BP image to obtain a lobe-free high-resolution sub-aperture BP image; Step 7: Traverse all sub-apertures and repeat steps 5 to 6 to coherently fuse all sub-aperture BP images to obtain a lobe-free, high-resolution SAR image.

2. The GCBP imaging method for non-orthogonal side-array multi-channel SAR according to claim 1, characterized in that, Constructing a non-frontal array multi-channel SAR geometric model, including: Taking the ground projection position of the radar array transmission channel at the center of the radar aperture as the origin, and the flight direction as... Establish a Cartesian coordinate system on the ground. Set the initial position of the high-speed mobile platform. , Platform height, velocity vector acceleration vector Furthermore, the direction of acceleration is the same as the direction of velocity, and the platform's dive angle Arbitrary scattering point in the imaging scene The coordinates are Establish a platform motion model; The signal from the radar array's transmission channel is equivalent to... The first equivalent phase center, the... The coordinates of the equivalent phase centers relative to the channel array center are: Define the channel direction tilt angle as The angle between the platform's dive trajectory direction and the channel arrangement direction is... Defined as: (1) when At this time, the conditions for non-positive side array observation are constituted; Arbitrary scattering point in the imaging scene To the The instantaneous slant distance of the phase center of each channel is expressed as: (2) in, For a slow time along the azimuth direction, the 4th order Maclaurin expansion of the slant range is as follows: (3) in, (4) Definition of the first The delay caused by the distribution distance between each channel and the reference channel for: (5) in, For the first The first channel and the first The projection of the spacing between the channels in the flight direction. This refers to the platform's flight speed.

3. The GCBP imaging method for non-orthogonal side array multi-channel SAR according to claim 1, characterized in that, Step 2, the process of pulse compression of the echo signal data received by the radar, includes: The radar baseband echo signal is obtained by down-converting the received radar echo signal data, and its expression is as follows: (6) in, For the fastest time along the distance direction, The slow time along the azimuth direction, The pulse width. To adjust the frequency, At the speed of light, For wavelength, For scattering points The backscattering coefficient; Based on the radar baseband echo signal, a range-direction system matching function is constructed, the expression of which is: (7) Based on the range-direction system matching function, the radar baseband echo signal is pulse-compressed. After pulse compression, the echo signal received by the reference channel phase center is: (8) in, For bandwidth.

4. The GCBP imaging method for non-frontal array multi-channel SAR according to claim 1, characterized in that, The calculation process for the weight vector of the reconstructed sub-aperture echo signal is as follows: Step 3.1: Calculate the Doppler center frequency of the platform. The calculation formula is as follows: (9) in, Let be the coefficients of the first-order term in the Taylor expansion of the slant distance history. Wavelength; Step 3.2: Based on the maximum Doppler frequency corresponding to the entire imaging grid and minimum Doppler frequency The Doppler bandwidth of the imaging scene is defined by the following expression: (10) Step 3.3: Based on the Doppler bandwidth of the imaging scene, fold the Doppler center frequency and search for all integers that satisfy the folding condition. This ensures that the center frequency of the folded Doppler falls within the Doppler bandwidth of the imaging scene. The folding condition is as follows: (11) in, The pulse repetition frequency; Define fuzzy weight vector All integers that satisfy the folding condition To form a vector: (12) in, and These are the smallest and largest fuzzy integers found, respectively. For imaging grid points The center of the fuzzy weight vector; Step 3.4: Based on the fuzzy weight vector center Constructing imaging grid points Multi-channel steering matrix at the location ; (13) Among them, matrix elements Defined as: (14) In non-frontal and diving situations, the phase term Based on the Doppler center frequency, the corrected expression is as follows: (15) in, For the first The delay caused by the distribution distance between each channel and the reference channel The angle between the platform's dive trajectory direction and the channel arrangement direction. It is an oblique perspective; respectively along The velocity vector in the direction; Step 3.5: Calculate the weight vector of the reconstructed sub-aperture echo signal, using the following formula: (16) in, It is a multi-channel steering matrix.

5. The GCBP imaging method for non-orthogonal side-array multi-channel SAR according to claim 1, characterized in that, The expression for the sidelobe suppression window function is as follows: (17) in, For imaging grid points The location center time, For the length of the window, , , All values ​​are window weights.

6. The GCBP imaging method for non-frontal array multi-channel SAR according to claim 1, characterized in that, The integration process of the BP imaging can be expressed as: (18) in, After pulse compression, the first Sub-aperture echo signals of each channel, For imaging grid points The corresponding two-way echo delay, This is the phase compensation term before BP integration. To reconstruct the weight vector of the sub-aperture echo signal, This is a sidelobe suppression window function. for At this moment Each channel to the imaging grid point Instantaneous slant distance, This is the slow time along the azimuth direction.

7. The GCBP imaging method for non-frontal array multi-channel SAR according to claim 1, characterized in that, The sub-aperture multi-channel BP image is subjected to two-stage spectral compression, azimuth upsampling, and despectral compression processing, including: Sub-aperture multichannel BP image with first-level spectral compression function Multiply to obtain the BP image after spectral support region alignment; Perform a range-direction FFT on the BP image after spectral support region alignment, and it becomes... Domain, and the second-order spectral compression function Multiplying them yields the BP image after spectral tilt correction; Perform azimuth FFT on the BP image after spectral tilt correction to obtain... In the wavenumber domain, wavenumber compression is completed, resulting in a spectrally compressed BP image. The BP image after spectral compression is zero-padding at both ends of the two-dimensional beam spectrum to complete azimuth upsampling; The upsampled BP image was subjected to two-stage despectral compression to obtain a high-resolution sub-aperture BP image without grating lobes.

8. The GCBP imaging method for non-frontal array multi-channel SAR according to claim 7, characterized in that, The expression for the two-stage spectral compression function is as follows: (20) in, For the range wave number, carrier frequency The corresponding wave number, At the speed of light, This serves as the virtual beam rotation center.

9. A GCBP imaging system for a non-frontal array multi-channel SAR of a hypersonic diving platform, characterized in that, include: Model building module: Constructs a non-orthogonal side array multi-channel SAR geometric model based on imaging scene information; Data segmentation module: Performs pulse compression and sub-aperture segmentation on the echo signal data received by the radar to obtain echo signals within multiple sub-apertures; Weight vector calculation module: For each grid point in the imaging area, construct a multi-channel steering matrix and calculate the weight vector of the reconstructed sub-aperture echo signal; Window function construction module: Construct a sidelobe suppression window function based on the effective synthetic aperture length; BP Imaging Module: Traverses all grid points in the imaging region, substitutes the weight vector and sidelobe suppression window function into the integral calculation of BP imaging, and performs BP imaging processing on the echo signal within the sub-aperture to obtain... Sub-aperture images of each channel, then Coherently accumulate the sub-aperture BP images of each channel to obtain a low-resolution sub-aperture multi-channel BP image without raster lobes. Despectral compression module: Performs two-stage spectral compression, azimuth upsampling and despectral compression processing on the sub-aperture multi-channel BP image in sequence to obtain a lobe-free high-resolution sub-aperture BP image; Image fusion module: Iterates through all sub-apertures and coherently fuses all sub-aperture BP images to obtain a lobe-free, high-resolution SAR image.

Images