Two-dimensional real-time back projection sensing imaging method, device, equipment and medium

Abstract

The invention discloses a two-dimensional real-time back projection sensing imaging method, device, equipment and medium based on a multi-antenna system, and the imaging method comprises the following steps: S1, carrying out Hilbert transform, frequency shift and fast Fourier transform on an echo signal, and outputting an echo signal frequency spectrum; s2, performing matched filtering on the sending signal and the echo signal, and outputting a time domain pulse compression result; s3, calculating the coordinates of the imaging area according to the imaging area and the grid size, and calculating the pixel signal intensity corresponding to the image through two-dimensional interpolation according to the pulse compression result; calculating a phase compensation factor by using the coordinates of the imaging area, performing phase compensation on an interpolation result, and finally outputting and pre-storing a two-dimensional range profile; and S4, windowing and superposing the multi-channel two-dimensional range profile, and outputting a two-dimensional SAR image in real time. Based on the SAR simulation imaging method, multi-antenna echo signals can be processed, the angle resolution can be improved, and real-time sensing imaging of the indoor environment can be achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of communication sensing imaging, and particularly relates to a two-dimensional real-time back-projection sensing imaging method, device, computer device and storage medium based on a multi-antenna system. Background Art

[0002] Communication Sensing (Integrated Sensing and Communication, ISAC) is a technology that deeply integrates communication and environmental sensing functions. By sharing spectrum resources and hardware platforms, it realizes the collaborative work of information transmission and target sensing. Among them, sensing imaging, as an important application, uses wireless signals to perform high-resolution modeling and real-time imaging of targets or environments, and is widely used in fields such as autonomous driving, smart cities, and industrial Internet of Things. Currently, the main sensing imaging methods are: Synthetic Aperture Radar (SAR) simulation imaging, communication signal passive sensing imaging, millimeter wave imaging, and compressive sensing imaging. SAR simulation imaging uses radio waves reflected by targets to generate high-resolution images through signal processing, and has all-weather adaptability and strong penetration ability; in addition, its real-time imaging ability has been fully verified in various application scenarios.

[0003] The SAR simulation imaging method is divided into two types: frequency domain and time domain. The frequency domain imaging method has high computational efficiency and is suitable for imaging wide scenes and large areas, especially widely used in remote sensing and ground monitoring. However, it is sensitive to orbital errors and motion parameters and requires precise geometric correction, otherwise it is easy to cause image distortion; in addition, it requires a linear aperture, and it is difficult to embed motion compensation, and the algorithms and correction steps are complex in some scenarios, increasing the difficulty of system development. The time domain imaging method uses a point-by-point imaging method, which can achieve high-resolution imaging and has no limitation on the size of the imaging scene, and is suitable for parallel hardware processing through digital signal processor devices; but a large number of interpolation operations are required for point-to-point image reconstruction, and the computational amount is large, especially when the target distance and processing area increase, the computational efficiency is significantly reduced.

[0004] The time domain back-projection algorithm (Back Projection Algorithm, BP) is an accurate SAR imaging method. This algorithm back-projects the echo signal after pulse compression for each pulse to the image domain, and performs coherent accumulation on pixel points, and finally reconstructs a high-resolution imaging result; however, the back-projection imaging method needs to perform two-dimensional interpolation at each pixel point in the imaging area in actual imaging, with high computational complexity and slow operation speed, especially with low efficiency when processing a large amount of data, and it is difficult to meet the requirements of high-precision and real-time imaging on traditional processor platforms (such as ARM, x86, etc.).

[0005] Field Programmable Gate Array (FPGA) has characteristics such as strong parallel processing ability, rich hardware resources, high programmability, low latency, and low power consumption. It integrates multipliers, RAM, and various arithmetic IP cores internally, and can significantly improve the real-time performance of imaging devices by using parallel processing and pipelining techniques. Implementing the back-projection imaging method on FPGA has several technical difficulties. First, the back-projection imaging method belongs to a computationally intensive task. A large number of multiplication and addition operations are required for FFT and two-dimensional interpolation. In real-time imaging applications, reasonably allocating computing resources is an important technical challenge. Second, the on-chip storage resources of FPGA are limited. The large data access requirements in the back-projection imaging method make storage management complex, and an efficient on-chip and off-chip data exchange and optimized cache mechanism are needed to reduce access latency. In addition, implementing the back-projection imaging method on FPGA also requires a trade-off between accuracy and latency to achieve real-time processing while meeting the operation accuracy requirements. Summary of the Invention

[0006] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a two-dimensional real-time back-projection sensing imaging method, device, computer device, and storage medium based on a multi-antenna system. This two-dimensional real-time back-projection sensing imaging method performs coherent superposition processing on the antenna echo signals at different positions to achieve high-resolution SAR imaging; and this two-dimensional real-time back-projection sensing imaging method can be deployed on an FPGA hardware platform and has the ability of real-time sensing imaging.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions: A two-dimensional real-time back-projection sensing imaging method based on a multi-antenna system, which is applied to a multi-antenna system. The multi-antenna system includes K antennas, K single channels, 1 analog-to-digital converter, and 1 two-dimensional back-projection sensing imaging module. Each antenna corresponds to 1 single channel; each antenna is used to receive electromagnetic wave signals from different directions to achieve omnidirectional signal acquisition of the target space; the single channel corresponding to the antenna processes the received radio frequency signal, including signal amplification, frequency conversion, and filtering, and converts the weak high-frequency radio frequency signal into an intermediate frequency signal with appropriate intensity suitable for subsequent processing; the analog-to-digital converter converts all the analog intermediate frequency signals output by the single channels into digital signals to achieve the conversion of the signal from the analog domain to the digital domain; the two-dimensional back-projection sensing imaging module uses the multiplexed digital signals output by the analog-to-digital converter to reconstruct the two-dimensional image of the target through the two-dimensional real-time back-projection sensing imaging method; the two-dimensional real-time back-projection sensing imaging method includes the following steps: S1. Perform Hilbert transform, frequency shifting, and fast Fourier transform on the echo signals received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; S2. Perform matched filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result; S3. Use a rectangular plane with a length of L meters and a width of W meters as the imaging area. This imaging area is evenly divided into I×J grid cells by a grid. I and J represent the number of grid cells in the length and width directions respectively. The size of each grid cell is determined by L / I and W / J, corresponding to the cell sizes in the length and width directions respectively. Calculate the imaging area coordinates based on the imaging area and the grid size, and calculate the signal intensity of the corresponding pixels of the image through two-dimensional interpolation according to the pulse compression result; then calculate the phase compensation factor based on the imaging area coordinates, and perform phase compensation on the interpolation result, and output the single-channel two-dimensional range image and store it in advance; S4. Window and stack multiple single-channel two-dimensional range images, and output the two-dimensional SAR image in real time.

[0008] Furthermore, the process of step S1 is as follows: S11. Perform Hilbert transform on the echo signal sequence received by the analog-to-digital converter to realize the conversion of the signal from the real number domain to the complex number domain, so as to extract the amplitude and phase information of the signal in the frequency domain analysis, and provide a basis for subsequent frequency translation and Fourier transform; S12. Perform frequency translation on the echo signal sequence after Hilbert transform to shift the echo signal frequency to the intermediate frequency band. The calculation formula is as follows:

[0009] where, is the intermediate frequency echo signal sequence of the kth antenna, is the echo signal sequence of the kth antenna, k = 1, 2,..., K, is the intermediate frequency, is the sampling frequency, n is the sampling point, is an imaginary number. This step simplifies the subsequent spectrum analysis process, reduces high-frequency noise interference, and improves the accuracy and efficiency of frequency domain processing; S13. Perform fast Fourier transform on the echo signal sequence after frequency translation, and output the spectrum of the echo signal sequence. This step converts the echo signal from the time domain to the frequency domain, and provides the necessary spectrum information for subsequent imaging and signal processing; Furthermore, the process of step S2 is as follows: S21. Perform Hilbert transform, frequency translation, windowing and fast Fourier transform on the transmitted signal sequence, and then store the spectrum sequence of the transmitted signal in the random access memory. This step avoids repeated calculation and redundant processing, effectively reduces the consumption of computing resources, and reduces the data processing delay of the system; S22. Perform complex conjugate multiplication of the spectrum sequence of the transmitted signal and the spectrum sequence of the echo signal to realize matched filtering. The calculation formula is as follows:

[0010] Among them, and are the real part and the imaginary part of the transmitted signal spectrum sequence, and are the real part and the imaginary part of the echo signal spectrum sequence, is an imaginary number. This step gives full play to the correlation in the frequency domain, accurately extracts the characteristics of the echo signal, and improves the accuracy and stability of target detection; S23. Perform an inverse fast Fourier transform on the frequency-domain pulse compression data obtained by matched filtering, and at the same time calculate the echo index, and output the time-domain pulse compression result and the echo index and store them in the random access memory. This step converts the frequency-domain pulse compression data to the time domain for processing, and adopts the time-domain backprojection imaging algorithm. Compared with backprojection imaging in the frequency domain, it greatly reduces the computational complexity and delay, and improves the real-time processing ability of the system.

[0011] Furthermore, the process of step S3 is as follows: S31. Calculate the imaging area coordinates according to the size of the imaging area and the grid size; subsequently, based on the position of the antenna and the coordinates of the imaging area, calculate the grid delay from each coordinate point to different antennas, and the calculation formula is as follows:

[0012] Among them, is the grid delay from the grid point at the i-th row and j-th column of the imaging area to the k-th antenna, and are the abscissa and ordinate of the grid point at the i-th row and j-th column in the imaging area, i = 1, 2,..., I, j = 1, 2,..., J, k = 1, 2,..., K, and are the abscissa and ordinate of the k-th antenna; S32. According to the grid delay of the coordinate points on the imaging area, read the adjacent echo index closest to the grid delay and the corresponding pulse compression result from the random access memory, and then calculate the signal intensity of the corresponding pixel of the image through two-dimensional interpolation , and the calculation formula is as follows:

[0013] Among them, for the k-th antenna, is the signal intensity corresponding to the grid point at the i-th row and j-th column of the imaging area, is the grid delay corresponding to the grid point at the i-th row and j-th column of the imaging area, and are the adjacent echo indices closest to the grid delay, and is the echo index corresponding to the k-th antenna and The pulse compression result of. This step accurately obtains the delay relationship between each grid point in the imaging area and the antenna, providing accurate delay information for subsequent signal processing; S33. Calculate the phase compensation factor according to the grid delay of the coordinate points on the imaging area. The calculation formula is as follows:

[0014] where, for the k-th antenna, is the phase compensation factor corresponding to the grid point at the i-th row and j-th column in the imaging area, is the carrier frequency, is the intermediate frequency, is the grid delay corresponding to the grid point at the i-th row and j-th column in the imaging area. This step accurately estimates the signal intensity of each grid point in the imaging area through a two-dimensional interpolation algorithm, ensuring the smooth transition and accuracy of the imaging quality, and avoiding distortion or blurring problems caused by data discreteness; Use the phase compensation factor to perform phase compensation on the pixel signal intensity output in step S32 to output the two-dimensional range image of the k-th antenna.

[0015] Furthermore, in step S4, for the single channels corresponding to K antennas, a Taylor window function with a length of K is set, and the two-dimensional range images output in step S3 are windowed and then accumulated to obtain a high-precision two-dimensional SAR image. This step effectively reduces the influence of sidelobes and reduces the interference of stray signals through the windowing process of the Taylor window for each antenna signal; by accumulating multi-channel signals, the angular resolution is effectively enhanced, avoiding the problem of insufficient resolution that may be caused by a single antenna signal, making the final two-dimensional SAR image more refined and realistic.

[0016] Furthermore, the echo signal sequences of each antenna are input serially in turn; for the echo signal sequence of each antenna, steps S1 to S3 are executed in a loop; in step S4, the two-dimensional range image of the current antenna is multiplied by the windowing coefficient, and then added to the historical two-dimensional range image, and the result is stored in the random access memory; after judging that a frame of image data is completed, the two-dimensional SAR image is output. The above steps significantly reduce the system's requirements for computing resources and storage space by serializing the processing of single-channel signals and only processing single-channel data each time; through the loop processing method, it not only ensures the efficient utilization of large-scale antenna data but also avoids the hardware resource pressure brought by high-parallel processing, thus effectively optimizing the computing efficiency.

[0017] In a second aspect, the present invention provides a two-dimensional real-time back-projection sensing imaging device for performing the above-mentioned two-dimensional real-time back-projection sensing imaging method. The two-dimensional real-time back-projection sensing imaging device includes: An echo signal processing module, configured to perform Hilbert transform, frequency shifting, and fast Fourier transform on the echo signals received by the analog-to-digital converters in the multi-antenna system, and output the echo signal spectrum; A signal matched filtering module, configured to perform matched filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result; A phase compensation module, configured to use a rectangular plane with a length of L meters and a width of W meters as the imaging area. The imaging area is evenly divided into I×J grid cells by a grid. I and J respectively represent the number of grids in the length and width directions. The size of each grid cell is determined by L / I and W / J, corresponding to the cell sizes in the length and width directions respectively. Calculate the imaging area coordinates according to the imaging area and grid size, and calculate the image corresponding pixel signal intensity through two-dimensional interpolation according to the pulse compression result; then calculate the phase compensation factor according to the imaging area coordinates, and perform phase compensation on the interpolation result according to the phase compensation factor, and respectively output and pre-store the single-channel two-dimensional range images; A two-dimensional SAR image output module, configured to window and superimpose multiple single-channel two-dimensional range images, and output the two-dimensional SAR image in real time.

[0018] In a third aspect, the present invention provides a computer device, including a processor and a memory for storing the executable program of the processor. When the processor executes the program stored in the memory, the above-mentioned two-dimensional real-time back-projection sensing imaging method is implemented.

[0019] In a fourth aspect, the present invention provides a storage medium storing a program, and when the program is executed by a processor, the above-mentioned two-dimensional real-time back-projection sensing imaging method is implemented.

[0020] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1. The present invention can process multi-antenna echo sequences, introduce windowing processing in pulse compression operations and coherent accumulation operations, and while improving the angular resolution, effectively suppress sidelobe interference in the imaging process; 2. The present invention can simultaneously perform grid delay calculation, phase compensation factor calculation, two-dimensional interpolation, and phase compensation operations. Among them, the phase compensation factor calculation operation and the two-dimensional interpolation operation can be executed in parallel, and a pipeline structure is adopted inside each operation, significantly accelerating the imaging speed. Through parallel design and pipeline mechanism, the data processing efficiency is effectively improved; 3. The present invention adopts a data processing algorithm based on coherent accumulation, which converts serial multi-antenna data into a form that can be processed in parallel. Only one radio frequency transceiver link and on-chip storage resources are required to complete the fast processing of multi-channel data, effectively reducing memory access latency and resource consumption, lowering hardware costs, and significantly improving the imaging speed. Description of the Drawings

[0021] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0022] Figure 1 It is a flowchart of the two-dimensional real-time back-projection sensing imaging method based on a multi-antenna system in Embodiment 1 of the present invention; Figure 2 It is a parallel pipelining processing block diagram in Embodiment 1 of the present invention; Figure 3 It is a Python simulation result diagram of back-projection sensing imaging in Embodiment 1 of the present invention; Figure 4 It is a result diagram of realizing back-projection sensing imaging by FPGA in Embodiment 1 of the present invention; Figure 5 It is a flowchart of the two-dimensional real-time back-projection sensing imaging method based on a multi-antenna system implemented under limited resource conditions in Embodiment 2 of the present invention; Figure 6 It is a parallel pipelining processing block diagram under limited resource conditions in Embodiment 2 of the present invention; Figure 7 It is a structural block diagram of the two-dimensional real-time back-projection sensing imaging device based on a multi-antenna system in Embodiment 3 of the present invention; Figure 8 It is a structural block diagram of the computer device in Embodiment 4 of the present invention. Detailed Embodiments

[0023] To enable those skilled in the art of the present technology to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0024] References to "embodiments" in this application mean that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of this application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described in this application can be combined with other embodiments.

[0025] Embodiment 1 As Figure 1 shown, this embodiment provides a two-dimensional real-time backprojection sensing imaging method based on a multi-antenna system, which is applied to a multi-antenna system. The multi-antenna system includes K antennas, K single channels, 1 analog-to-digital converter, and 1 two-dimensional backprojection sensing imaging module. Each antenna corresponds to 1 single channel; each antenna is used to receive electromagnetic wave signals from different directions; the single channel corresponding to the antenna processes the received radio frequency signal, including signal amplification, frequency conversion, and filtering, and converts the high-frequency radio frequency signal into an intermediate-frequency signal; the analog-to-digital converter converts the analog intermediate-frequency signals output by all single channels into digital signals; the two-dimensional backprojection sensing imaging module uses the multiplexed digital signals output by the analog-to-digital converter to reconstruct the two-dimensional image of the target through the two-dimensional real-time backprojection sensing imaging method, including the following steps: S0. Determine the system parameters of the backprojection imaging method module. Among them, the multi-antenna system needs to determine parameters. The system parameters include operating clock, input and output data types and lengths, sampling frequency, intermediate-frequency frequency, carrier frequency, receiving antenna coordinates, number of antennas, and transmission signal sequence types and lengths, etc. Specifically, it includes the following steps: S01. Set the operating clock to 100 MHz, and the sampling frequency is 2.4576 GHz, the intermediate-frequency frequency is 614.4 MHz, the carrier frequency is 7.425 GHz, and the signal bandwidth is 400 MHz; S02. The echo signal output from the ADC module is the input of the backprojection imaging method. The input of this module is a real number sequence with a length of 4096 points and a data type of 16-bit signed fixed-point number; the transmission signal sequence is a real number sequence with a length of 2472 points and a data type of 16-bit signed fixed-point number; in this embodiment, the length L of the imaging area is 4 meters, and the width W is 4 meters. This area is evenly divided into I×J grid cells through grid division, where the number of grid cells I in the length direction is 300, and the number of grid cells J in the width direction is 300. The size of the output SAR image is , and the data length output by this module is 90000 points, and the data type is 32-bit signed floating-point number.

[0026] S1. Perform Hilbert transform, frequency translation, and fast Fourier transform on the echo signal received by the analog-to-digital converter, and output the echo signal spectrum. The specific steps are as follows: S11. Input the echo signal sequence output by the ADC module, change the sequence data type to 16-bit signed fixed-point number, and the data length is 4096; S12. Perform zero-padding on the echo signal sequence in step S11, fill 4096 zero-value data at the end of the sequence, and extend the sequence length to 8192 points; S13. Call FDATool in MATLAB to design the Hilbert filter coefficients. The response type is set to Hilbert, the filter order is set to 32, the sampling frequency is set to 2.4576 GHz, and the filter passband range is from 61.44 MHz to 1167.36 MHz, which can process signals in the frequency range from 414.4 MHz to 814.4 MHz in this embodiment; use FDATool to automatically generate 32-order Hilbert filter coefficients and convert them into 16-bit signed fixed-point numbers, and then output them to the coe file; S14. Call the FIR IP core in Xilinx IP Catalog, import the coe file in step S13 for the filter coefficients of the IP core, set the filter type to Hilbert, the clock frequency to 100 MHz, the input data type to 16-bit signed fixed-point number, set the quantization type of the filter coefficients to Maximize Dynamic Range, and the real and imaginary parts of the output data are 16-bit signed integers; S15. Input the zero-padded echo signal sequence in step S12 into the FIR IP core to complete the Hilbert transform, and output a complex sequence with a data type of 16-bit signed integer; S16. Call the Complex Multiplier IP core in Xilinx IP Catalog, input the complex sequence of the echo signal and the frequency translation coefficient sequence in step S25, and the calculation formula is as follows:

[0027] where, is the intermediate-frequency echo signal sequence of the k-th antenna, is the echo signal sequence of the k-th antenna, k = 1, 2,..., K, is the intermediate frequency, is the sampling frequency, n is the sampling point, is an imaginary number, and the output data type is a 32-bit signed integer complex sequence with a length of 8192; S17. Call the FFT IP core in the Xilinx IP Catalog. Set the working mode of the IP core to Pipelined, Streaming I / O data stream mode, with a processing length of 8192 points. The input data type is 32-bit signed fixed-point complex numbers, and the output data type is 32-bit signed fixed-point complex numbers. S18. Input the complex number sequence in step S16 into the FFT IP core. The configuration parameters of this IP core are to perform FFT, complete the fast Fourier transform, and output the echo signal spectrum sequence with a length of 8192 points and a data type of 32-bit signed fixed-point complex numbers.

[0028] S2. Perform matched filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result, which specifically includes the following steps: S21. In this embodiment, the transmitted signal sequence is a linear frequency modulation continuous wave (FMCW) with a bandwidth of 400 MHz. Call the FDATool in MATLAB to design a Hilbert filter and perform Hilbert transform on the transmitted signal. Use MATLAB to shift the frequency of the transmitted signal. The specific calculation formula is:

[0029] where, is the transmitted signal sequence, is the intermediate frequency, is the sampling frequency, n is the sampling point, is the imaginary number; Call the taylorwin function in MATLAB to window the frequency-shifted transmitted signal sequence, and call the fft function in MATLAB to perform fast Fourier transform on the windowed transmitted signal sequence, and output a complex number sequence with a length of 8192 points and a data type of 32-bit signed fixed-point numbers. This sequence represents the spectrum of the transmitted signal sequence. S22. Call the Complex Multiplier IP core in the Xilinx IP Catalog. Input the negative of the imaginary part of the transmitted signal spectrum in step S31 into the IP core, and then input the echo signal spectrum sequence in step S18 into the IP core. Perform complex conjugate dot multiplication on the spectrum sequence of the transmitted signal and the spectrum sequence of the echo signal to achieve matched filtering. The calculation formula is as follows:

[0030] where, and are the real part and the imaginary part of the transmitted signal spectrum sequence, and are the real part and the imaginary part of the echo signal spectrum sequence, is an imaginary number. Call the Floating-point IP core in the Xilinx IP Catalog to convert the data type of the pulse compression result to a 32-bit signed floating point number; S23. Calculate the echo index in MATLAB, and the calculation formula is as follows:

[0031] where N is the length 8192 of the echo signal spectrum sequence, n is the sampling point, is the sampling frequency; then convert the sequence into a 16-bit signed fixed-point number and convert it into a.coe file; in the FPGA, call the Block Memory Generator IP core in the Xilinx IP Catalog, set the memory type to single-port ROM, and store the echo index sequence in this single-port ROM; S24. Input the complex sequence in step S22 into the FFT IP core. The configuration parameters of this IP core are to perform IFFT, and the output data type is 32-bit signed floating point number to complete the inverse fast Fourier transform; call the Block Memory Generator IP core in the Xilinx IP Catalog, set the memory type to single-port RAM, and store the pulse compression result in this single-port RAM.

[0032] S3. Calculate the imaging area grid coordinate parameters according to the imaging area and the grid size, calculate the imaging area grid delay and phase compensation coefficient according to the imaging area grid coordinate parameters, and perform two-dimensional interpolation and phase compensation on the imaging area according to the pulse compression result. As Figure 2 shown, step S3 includes grid delay calculation operation, phase compensation factor calculation operation, two-dimensional interpolation operation and phase compensation operation. In this embodiment, step S3 is processed in parallel and pipelined, and specifically includes the following steps: S31. The receiving antenna in this embodiment is a Uniform Linear Array Antenna (ULA), the number of antennas K is 64, the geometric center of the antenna array is used as the origin of the coordinate system, and the arrangement direction of the antenna array is defined as the Y-axis direction; the size of the imaging area is 4m ×4m, and the SAR image resolution is ; S32. Use MATLAB to calculate the X parameter and Y parameter of the imaging area coordinates. In this embodiment, the arrangement direction of the antenna array is defined as the Y-axis direction. For the kth antenna, k = 1, 2,..., 64, the calculation formula of the X parameter is:

[0033] Among them, i and j are the row number and column number corresponding to the grid points in the imaging area, i = 1, 2, …, 300, j = 1, 2, …, 300; the calculation formula of the Y parameter is:

[0034] Among them, i and j are the row number and column number corresponding to the grid points in the imaging area, i = 1, 2, …, 300, j = 1, 2, …, 300, is the y-axis coordinate of the k-th antenna element, k = 1, 2, …, 64; Convert the X parameter and Y parameter into 32-bit signed floating-point numbers; for the convenience of parallel pipelining processing, as Figure 2 shown, store the X parameter in the Flip-Flop resource in the array structure of the FPGA, with a data length of 300; store the Y parameter in the single-port ROM, with a data length of 19,200; S33. Execute 8 times of grid delay calculation operations in parallel. For the k-th antenna, k = 1, 2, …, 64, calculate the grid delay between each grid point and the receiving antenna element. The specific implementation steps of this operation are: S331. For the k-th antenna, k = 1, 2, …, 64, take out the corresponding parameters of the grid at the i-th row and j-th column from the X parameter array and the Y parameter RAM and , i = 1, 2, …, 300, j = 1, 2, …, 300; S332. Calculate the grid delay, and the calculation formula is as follows:

[0035] Among them, k is the antenna serial number, k = 1, 2, …, 64, i and j are the row number and column number corresponding to the grid points in the imaging area, c is the speed of light, and the value is ; S34. Execute 8 times of two-dimensional interpolation operations and phase compensation factor calculation operations in parallel. The two-dimensional interpolation operation can calculate the signal intensity of the corresponding pixel in the image. The specific implementation steps of this operation are: According to the grid delay in step S33, read the adjacent echo index closest to the grid delay and the corresponding pulse compression result from the RAM, and then calculate the signal intensity of the corresponding pixel in the image through two-dimensional interpolation. The calculation formula is as follows:

[0036] Among them, for the k-th antenna, is the signal intensity corresponding to the grid point at the i-th row and j-th column in the imaging area, is the grid delay corresponding to the grid point at the i-th row and j-th column of the imaging region, and is the adjacent echo index closest to the grid delay, and is the echo index corresponding to the k-th antenna and is the pulse compression result; The phase compensation factor calculation operation calculates the phase compensation factor according to the grid delay of the coordinate points on the imaging region. The calculation formula is as follows:

[0037] Among them, for the k-th antenna, is the phase compensation factor corresponding to the grid point at the i-th row and j-th column of the imaging region, is the carrier frequency, is the intermediate frequency, is the grid delay corresponding to the grid point at the i-th row and j-th column of the imaging region, is the imaginary number; S35. Execute 8 phase compensation operations in parallel, and use the compensation parameters calculated by the phase compensation factor calculation operation to perform phase compensation on the pixel signal intensity output by the two-dimensional interpolation module, and output the two-dimensional range image of the k-th antenna. The data length is 90000, and the data type is 32-bit signed floating-point number.

[0038] S4. Call the Block Memory Generator IP core in the Xilinx IP Catalog in the FPGA. Set the memory type to dual-port RAM, the RAM width to 64 bits, and the depth to 90000 for temporarily storing the windowed two-dimensional range image; for K antennas, set a Taylor window function with a length of K, and perform windowing and accumulation on the two-dimensional range image output in step S3 to obtain a high-precision two-dimensional SAR image.

[0039] The following are the test environment and specific parameter settings of this embodiment: In the indoor area, divide of the imaging region, place a corner reflector at the center of the imaging region. The receiving antenna of this embodiment is a uniform linear array antenna. Take the geometric center of the antenna array as the origin of the coordinate system, define the arrangement direction of the antenna array as the Y-axis direction, the antenna spacing is half a wavelength, and the direction perpendicular to the arrangement direction of the antenna array is defined as the X-axis direction. The transmitting antenna is placed at the coordinate origin of the imaging region; The specific parameter settings are: sampling frequency = 2.4576 GHz, intermediate frequency = 614.4 MHz, carrier frequency = 7.425 GHz, signal bandwidth B = 400 MHz, number of antennas ; Deploy the imaging module of the two-dimensional backprojection imaging method on the FPGA. Transmit an FMCW signal with a bandwidth of 400 MHz through the transmitting antennas. The length of the transmitted signal sequence is 2472 points. While transmitting the signal, switch the antennas and the corresponding single channels in sequence. The 64-channel echo signals are input into the imaging module of the two-dimensional backprojection imaging method in a serial manner. After the module generates the two-dimensional SAR image data, it is transmitted to the PC through Ethernet for reconstructing the two-dimensional environmental image; Figure 3 and Figure 4 respectively show the Python simulation results and FPGA implementation results of the two-dimensional backprojection imaging method. Figure 3 It is the result obtained by using Python to implement the two-dimensional backprojection imaging algorithm and through simulation testing for collecting multi-antenna echo signals from a multi-antenna system; Figure 4 is the result obtained after executing the two-dimensional backprojection imaging method on the FPGA according to the specific steps of the invention embodiment.

[0040] The correct imaging result should be reflected in that the energy of the SAR image is concentrated on the corner reflector in the center of the imaging area. In Figure 3 , the Python simulation results show that the energy is accurately focused on the corner reflector in the center of the imaging area, indicating high imaging accuracy and accurate positioning in the simulation test; in Figure 4 , the result implemented on the FPGA also shows the energy focusing, which is consistent with the theoretical expectation.

[0041] By comparing Figure 3 and Figure 4 it can be found that the imaging results of both can concentrate the energy on the corner reflector in the center of the imaging area, which shows that the two-dimensional backprojection imaging method implemented on the FPGA meets the theoretical requirements in terms of accuracy and positioning. In addition, this method can not only accurately sense the environment in the imaging area and reconstruct the environmental image, but also meet the requirements of real-time imaging, further verifying its reliability and practicality in practical applications. Embodiment 2 Such as Figure 5As shown in the figure, this embodiment provides a two-dimensional real-time back-projection sensing imaging method based on a multi-antenna system under limited resource conditions. Compared with Embodiment 1, in this embodiment, by optimizing the design of the grid delay parameter, it is not necessary to calculate the adjacent echo index closest to the grid delay and perform division operations during two-dimensional interpolation, and it is not necessary to use a single-port ROM to store the echo index data. Compared with Embodiment 1, this method saves the hardware resources required for 8 32-bit signed floating-point subtractors, 8 32-bit signed floating-point dividers, and a single-port ROM with a length of 8192 and a depth of 32 bits. This embodiment specifically includes the following steps: S0. Determine the system parameters of the back-projection imaging method module. The system parameters include the module operating clock, input and output data types and lengths, sampling frequency, intermediate frequency, carrier frequency, receiving antenna coordinates, number of antennas, and transmission signal sequence type and length, etc. For the specific implementation, refer to step S0 of Embodiment 1.

[0042] S1. Perform Hilbert transform, frequency shift, and fast Fourier transform on the echo signal received by the analog-to-digital converter, and output the echo signal spectrum. For the specific implementation steps, refer to step S1 of Embodiment 1.

[0043] S2. Perform matched filtering on the transmission signal and the echo signal, and output the time-domain pulse compression result. The specific implementation steps are as follows: S21. In this embodiment, the transmission signal sequence is a linear frequency modulation continuous wave (FMCW) with a bandwidth of 400 MHz. Call the FDATool in MATLAB to design a Hilbert filter and perform Hilbert transform on the transmission signal, and use MATLAB to perform frequency shift on the transmission signal. The specific calculation formula is:

[0044] where, is the transmission signal sequence, is the intermediate frequency, is the sampling frequency, n is the sampling point, is an imaginary number; call the taylorwin function in MATLAB to window the frequency-shifted transmission signal sequence, and call the fft function in MATLAB to perform fast Fourier transform on the windowed transmission signal sequence, and output a complex sequence with a length of 8192 points and a data type of 32-bit signed fixed-point number. This sequence represents the transmission signal sequence spectrum; S22. Call the Complex Multiplier IP core in the Xilinx IP Catalog, take the negative of the imaginary part of the transmitted signal spectrum in step S31) and input it into the IP core, then input the echo signal spectrum sequence in step S18 into the IP core, and perform complex conjugate dot multiplication on the spectrum sequence of the transmitted signal and the spectrum sequence of the echo signal to achieve matched filtering. The calculation formula is as follows:

[0045] Wherein, and are the real part and the imaginary part of the transmitted signal spectrum sequence, and are the real part and the imaginary part of the echo signal spectrum sequence, is an imaginary number. Call the Floating-point IP core in the Xilinx IP Catalog to convert the data type of the pulse compression result to 32-bit signed floating-point number; S3. Calculate the imaging area grid coordinate parameters according to the imaging area and the grid size, then design the imaging area grid delay and phase compensation coefficients applicable to the limited resource conditions according to the imaging area grid coordinate parameters, and then perform two-dimensional interpolation on the imaging area according to the pulse compression result and compensate the phase to output a single-channel two-dimensional range image. Figure 6 The following figure shows the schematic diagram of the parallel pipelining process of step S3 in this embodiment. This step includes grid delay calculation operation, phase compensation factor calculation operation, two-dimensional interpolation operation and phase compensation operation, which specifically include the following steps: S31. Refer to step S31 of Embodiment 1; S32. Use MATLAB to calculate the X parameter and Y parameter of the imaging area coordinates. In this embodiment, the arrangement direction of the antenna array is defined as the Y-axis direction. For the k-th antenna, k = 1, 2,..., 64, the calculation formula of the X parameter is:

[0046] Wherein, i and j are the row number and column number corresponding to the imaging area grid points, i = 1, 2,..., 300, j = 1, 2,..., 300; the calculation formula of the Y parameter is:

[0047] Wherein, i and j are the row number and column number corresponding to the imaging area grid points, i = 1, 2,..., 300, j = 1, 2,..., 300, is the y-axis coordinate of the k-th antenna element, k = 1, 2,..., 64; S33. Design the imaging area grid delay parameters DX and DY applicable to the limited resource conditions according to the X parameter and Y parameter in step S32. For the k-th antenna, k = 1, 2, …, 64, the calculation formula of DX is as follows:

[0048] where DX is the abscissa parameter of the imaging area grid delay applicable to the limited resource conditions, and X is the abscissa parameter of the imaging area grid coordinates obtained in step S32, is the sampling frequency; The calculation formula of DY is as follows:

[0049] where DY is the ordinate parameter of the imaging area grid delay applicable to the limited resource conditions, and Y is the ordinate parameter of the imaging area grid coordinates obtained in step S32, is the sampling frequency; Convert the DX parameter and the D parameter into 32-bit signed floating-point numbers; for the convenience of parallel pipelining processing, as Figure 6 shown, store the DX parameter in the Flip-Flop resource in the array structure of the FPGA, with a data length of 300; store the DY parameter in a single-port ROM, with a data length of 19,200; S34. Execute 8 times of grid delay calculation operations in parallel. For the k-th antenna, k = 1, 2, …, 64, calculate the grid delay between each grid point and the receiving antenna unit under the limited resource conditions. The specific implementation steps of this operation are as follows: S341. For the k-th antenna, k = 1, 2, …, 64, take out the grid corresponding parameters of the i-th row and j-th column from the DX parameter array and the DY parameter RAM and , i = 1, 2, …, 300, j = 1, 2, …, 300; S342. Calculate the grid delay under the limited resource conditions. The calculation formula is as follows:

[0050] where k is the antenna serial number, k = 1, 2, …, 64, i and j are the row number and column number corresponding to the imaging area grid point, i = 1, 2, …, 300, j = 1, 2, …, 300, is the grid delay under the limited resource conditions, c is the speed of light, and the value is ; S35. Execute 8 times of two-dimensional interpolation operations and phase compensation factor calculation operations in parallel. The two-dimensional interpolation operation can calculate the signal intensity of the corresponding pixel of the image. The specific implementation steps of this operation are as follows: According to the grid delay under limited resource conditions in step S41 , the echo index calculation formula is as follows:

[0051] Take out the pulse compression results corresponding to addresses m and m + 1 from the RAM storing the pulse compression sequence, and then obtain the signal intensity of the corresponding pixel in the image through two-dimensional interpolation calculation under limited resource conditions. The calculation formula is as follows:

[0052] Among them, for the k-th antenna, is the signal intensity corresponding to the grid point at the i-th row and j-th column in the imaging area, is the grid delay of the grid point at the i-th row and j-th column in the imaging area, i = 1, 2,..., 300, j = 1, 2,..., 300, k = 1, 2,..., 64, and are the pulse compression results corresponding to addresses m and m + 1 in the RAM storing the pulse compression sequence for the k-th antenna; S36. According to the grid delay under limited resource conditions , calculate the phase compensation factor. The calculation formula is as follows:

[0053] Among them, for the k-th antenna, is the phase compensation factor corresponding to the grid point at the i-th row and j-th column in the imaging area, i = 1, 2,..., 300, j = 1, 2,..., 300, k = 1, 2,..., 64, is the carrier frequency, is the intermediate frequency, is the sampling frequency, is the grid delay under limited resources corresponding to the grid point at the i-th row and j-th column in the imaging area, is an imaginary number; S37. Execute 8 phase compensation operations in parallel, and use the compensation parameters calculated by the phase compensation factor calculation operation to perform phase compensation on the pixel signal intensity output by the two-dimensional interpolation module, and output the two-dimensional range image of the k-th antenna. The data length is 90000, and the data type is 32-bit signed floating-point number.

[0054] S4. Call the Block Memory Generator IP core in the Xilinx IP Catalog in the FPGA. Set the memory type to dual-port RAM, the RAM width to 64 bits, and the depth to 90,000 for temporarily storing the two-dimensional range image after windowing. For K antennas, set a Taylor window function with a length of K, window and accumulate the two-dimensional range image output in step S3) to obtain a high-precision two-dimensional SAR image.

[0055] Embodiment 4 Referring to Figure 7 , this embodiment provides a two-dimensional real-time back-projection sensing imaging device based on a multi-antenna system. The two-dimensional real-time back-projection sensing imaging device based on a multi-antenna system includes an echo signal processing module 701, a signal matching and filtering module 702, a phase compensation module 703, and a two-dimensional SAR image output module 704 that are sequentially and connected in sequence, where: The echo signal processing module 701 is configured to perform Hilbert transform, frequency shifting, and fast Fourier transform on the echo signals received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; The signal matching and filtering module 702 is configured to perform matching filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result; The phase compensation module 703 is configured to use a rectangular plane with a length of L meters and a width of W meters as the imaging area. The imaging area is evenly divided into I×J grid cells by a grid. I and J respectively represent the number of grid cells in the length and width directions. The size of each grid cell is determined by L / I and W / J, corresponding to the cell sizes in the length and width directions respectively. Calculate the imaging area coordinates according to the imaging area and the grid size, and calculate the image corresponding pixel signal intensity through two-dimensional interpolation according to the pulse compression result; then calculate the phase compensation factor according to the imaging area coordinates, and perform phase compensation on the interpolation result according to the phase compensation factor, and respectively output and pre-store the single-channel two-dimensional range image; The two-dimensional SAR image output module 704 is configured to window and superimpose multiple single-channel two-dimensional range images, and output a two-dimensional SAR image in real time.

[0056] The content in the above method embodiments is applicable to this system embodiment, and the functions specifically implemented in this system embodiment are the same as those in the above method embodiments.

[0057] Embodiment 4 This embodiment provides a computer device, which can be a computer, such as Figure 8As shown in the figure, it includes a processor 802, a memory, an input device 803, a display 804, and a network interface 805 connected through a system bus 801. The processor is used to provide computing and control capabilities. The memory includes a non-volatile storage medium 806 and an internal memory 807. The non-volatile storage medium 806 stores an operating system, a computer program, and a database. The internal memory 807 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the processor 802 executes the computer program stored in the memory, it implements a two-dimensional real-time backprojection sensing imaging method based on a multi-antenna system proposed in Embodiment 1 above. The two-dimensional real-time backprojection sensing imaging method based on a multi-antenna system includes the following steps: S1. Perform Hilbert transform, frequency translation, and fast Fourier transform on the echo signal received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; S2. Perform matched filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result; S3. Use a rectangular plane with a length of L meters and a width of W meters as the imaging area. The imaging area is evenly divided into I×J grid cells by a grid. I and J respectively represent the number of grids in the length and width directions. The size of each grid cell is determined by L / I and W / J, corresponding to the cell size in the length and width directions respectively. Calculate the imaging area coordinates according to the imaging area and the grid size, and calculate the image corresponding pixel signal intensity through two-dimensional interpolation according to the pulse compression result; then calculate the phase compensation factor according to the imaging area coordinates, and perform phase compensation on the interpolation result according to the phase compensation factor, and output and pre-store the single-channel two-dimensional range image respectively; S4. Window and superimpose multiple single-channel two-dimensional range images, and output a two-dimensional SAR image in real time.

[0058] Embodiment 5 This embodiment provides a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, it implements a two-dimensional real-time backprojection sensing imaging method based on a multi-antenna system proposed in Embodiment 1 above. The two-dimensional real-time backprojection sensing imaging method based on a multi-antenna system includes the following steps: S1. Perform Hilbert transform, frequency translation, and fast Fourier transform on the echo signal received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; S2. Perform matched filtering on the transmitted signal and the echo signal, and output the time-domain pulse compression result; S3. Take a rectangular plane with a length of L meters and a width of W meters as the imaging area. This imaging area is evenly divided into I×J grid cells by a grid. I and J respectively represent the number of grids in the length and width directions. The size of each grid cell is determined by L / I and W / J, corresponding to the cell sizes in the length and width directions respectively. Calculate the coordinates of the imaging area based on the imaging area and grid size, and calculate the pixel signal intensity of the image corresponding to the pulse compression result through two-dimensional interpolation; then calculate the phase compensation factor using the imaging area coordinates, perform phase compensation on the interpolation result according to the phase compensation factor, and output and pre-store the single-channel two-dimensional range image respectively. S4. Window and superimpose multiple single-channel two-dimensional range images, and output the two-dimensional SAR image in real time.

[0059] In this embodiment, the storage medium described may be a magnetic disk, an optical disc, a computer memory, a random access memory (RAM), a USB flash drive, a mobile hard disk, or other media.

[0060] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously.

[0061] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combinations of these technical features do not conflict, they should be considered as the scope described in this specification.

[0062] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A two-dimensional real-time back projection perception imaging method based on a multi-antenna system, applied to a multi-antenna system, the multi-antenna system includes K antennas, K single channels, an analog-to-digital converter and a two-dimensional back projection perception imaging module, each antenna corresponds to a single channel; each antenna is used to receive electromagnetic wave signals from different directions; the single channel corresponding to the antenna processes the received radio frequency signal, including signal amplification, frequency conversion and filtering, and converts the high-frequency radio frequency signal into an intermediate frequency signal; the analog-to-digital converter converts the analog intermediate frequency signals output by all single channels into digital signals; the two-dimensional back projection perception imaging module uses the multi-channel digital signals output by the analog-to-digital converter to reconstruct the two-dimensional image of the target through the two-dimensional real-time back projection perception imaging method, characterized in that The two-dimensional real-time back-projection sensing imaging method comprises the following steps: S1. Perform Hilbert transform, frequency shifting and fast Fourier transform on the echo signal received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; S2, performing matched filtering on the transmission signal and the echo signal, and outputting the time domain pulse compression result; S3, a rectangular plane with a length of L meters and a width of W meters is used as an imaging area, and the imaging area is evenly divided into I×J grid units through a grid, where I and J represent the number of grids in the length and width directions, respectively. The size of each grid unit is determined by L / I and W / J, corresponding to the unit size in the length and width directions, respectively. The coordinates of the imaging area are calculated according to the imaging area and the grid size, and the signal intensity of the corresponding pixel of the image is calculated by two-dimensional interpolation according to the pulse compression result; then the phase compensation factor is calculated using the coordinates of the imaging area, and the interpolation result is phase compensated according to the phase compensation factor, and the single-channel two-dimensional range image is output and pre-stored respectively; S4. Windowing and superimposing multiple single-channel two-dimensional range images to output a two-dimensional SAR image in real time.

2. The two-dimensional real-time back-projection perception imaging method according to claim 1, characterized in that: The process of step S1 is as follows: S11, performing Hilbert transform on the echo signal sequence received by the analog-to-digital converter to achieve conversion of the signal from the real domain to the complex domain; S12, frequency shifting is performed on the echo signal sequence after Hilbert transformation, and the echo signal frequency is shifted to the intermediate frequency band. The calculation formula is as follows: in, is the intermediate frequency echo signal sequence of the kth antenna, is the echo signal sequence of the kth antenna, k=1,2,…,K, is the intermediate frequency, is the sampling frequency, n is the sampling point, is an imaginary number; S13, performing fast Fourier transform on the frequency-shifted echo signal sequence, and outputting the frequency spectrum of the echo signal sequence.

3. The two-dimensional real-time back-projection perception imaging method according to claim 1, characterized in that: The process of step S2 is as follows: S21, performing Hilbert transform, frequency shifting, windowing and fast Fourier transform on the transmission signal sequence, and then storing the spectrum sequence of the transmission signal into a random access memory; S22, perform complex conjugate point multiplication on the spectrum sequence of the transmitted signal and the spectrum sequence of the echo signal to implement matched filtering. The calculation formula is as follows: in, and are the real and imaginary parts of the transmitted signal spectrum sequence, and are the real and imaginary parts of the echo signal spectrum sequence, is an imaginary number; S23, performing inverse fast Fourier transform on the frequency domain pulse compression data obtained by matched filtering, calculating the echo index at the same time, outputting the time domain pulse compression result and the echo index and storing them in a random access memory.

4. The two-dimensional real-time back-projection perception imaging method according to claim 1, characterized in that: The process of step S3 is as follows: S31, calculating the coordinates of the imaging area according to the size of the imaging area and the grid size; then, based on the position of the antenna and the coordinates of the imaging area, calculating the grid delay from each coordinate point to different antennas, the calculation formula is as follows: in, is the grid delay from the grid point in the i-th row and j-th column of the imaging area to the k-th antenna, and are the horizontal and vertical coordinates of the grid point in the i-th row and j-th column in the imaging area, i=1,2,…,I, j=1,2,…,J, k=1,2,…,K, and are the abscissa and ordinate of the kth antenna; S32, according to the grid delay of the coordinate point on the imaging area, read the adjacent echo index closest to the grid delay and the corresponding pulse compression result from the random access memory, and then calculate the corresponding pixel signal intensity of the image by two-dimensional interpolation , the calculation formula is as follows: Among them, for the kth antenna, is the signal intensity corresponding to the grid point in the i-th row and j-th column of the imaging area, is the grid delay corresponding to the grid point in the i-th row and j-th column of the imaging area, and is the index of the neighboring echo closest to the grid delay, and is the echo index corresponding to the kth antenna and The pulse compression result of S33, calculating the phase compensation factor according to the grid delay of the coordinate points on the imaging area, the calculation formula is as follows: Among them, for the kth antenna, is the phase compensation factor corresponding to the grid point in the i-th row and j-th column of the imaging area, is the carrier frequency, is the intermediate frequency, is the grid delay corresponding to the grid point in the i-th row and j-th column of the imaging area; The pixel signal intensity output in step S32 is adjusted by using the phase compensation factor. Perform phase compensation and output the two-dimensional range image of the kth antenna.

5. The two-dimensional real-time back-projection perception imaging method according to claim 1, characterized in that: In the step S4, a Taylor window function with a length of K is set for a single channel corresponding to the K antennas, and the two-dimensional range image outputted in the step S3 is windowed and then accumulated to obtain a high-precision two-dimensional SAR image.

6. The two-dimensional real-time back-projection sensing imaging method according to claim 1, characterized in that: The echo signal sequences of each antenna are input serially in sequence; for the echo signal sequence of each antenna, steps S1 to S3 are executed cyclically; In step S4, the two-dimensional range image of the current antenna is multiplied by the windowing coefficient, and then added to the historical two-dimensional range image, and the result is stored in the random access memory; after judging that one frame of image data is completed, the two-dimensional SAR image is output.

7. A two-dimensional real-time back-projection perception imaging device, used to execute the two-dimensional real-time back-projection perception imaging method according to any one of claims 1 to 7, characterized in that: The two-dimensional real-time back-projection perception imaging device comprises: The echo signal processing module is used to perform Hilbert transform, frequency shifting and fast Fourier transform on the echo signal received by the analog-to-digital converter in the multi-antenna system, and output the echo signal spectrum; A signal matching filter module is used to match filter the transmission signal and the echo signal and output the time domain pulse compression result; The phase compensation module is used to take a rectangular plane with a length of L meters and a width of W meters as an imaging area. The imaging area is evenly divided into I×J grid units through a grid, where I and J represent the number of grids in the length and width directions, respectively. The size of each grid unit is determined by L / I and W / J, which correspond to the unit size in the length and width directions, respectively. The coordinates of the imaging area are calculated according to the imaging area and the grid size, and the signal intensity of the corresponding pixel of the image is calculated through two-dimensional interpolation according to the pulse compression result; then the phase compensation factor is calculated using the coordinates of the imaging area, and the interpolation result is phase compensated according to the phase compensation factor, and the single-channel two-dimensional range image is output and pre-stored respectively; The two-dimensional SAR image output module is used to window and superimpose multiple single-channel two-dimensional range images and output two-dimensional SAR images in real time.

8. A computer device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, the two-dimensional real-time back-projection perception imaging method described in any one of claims 1 to 6 is implemented.

9. A storage medium storing a program, characterized in that: When the program is executed by a processor, the two-dimensional real-time back-projection perception imaging method described in any one of claims 1 to 6 is implemented.

Images