A handheld high-resolution millimeter wave synthetic aperture radar imaging method

Abstract

The application discloses a handheld high-resolution millimeter wave synthetic aperture radar imaging method, which comprises the following steps: mixing the signal collected by the handheld high-resolution millimeter wave radar to obtain an intermediate frequency signal; performing fast Fourier transform processing and residual video phase removal on the intermediate frequency signal; calculating the frequency response corresponding to the distance between the radar antenna phase center and the pixel point through linear interpolation; obtaining a preliminary result according to backward projection imaging; calculating position error estimation according to a minimum entropy self-focusing algorithm; correcting the trajectory according to the position error estimation, and then re-executing the backward projection imaging. The minimum entropy self-focusing backward projection algorithm provided by the application takes the minimization of image entropy as an optimization target, compensates the position error of the radar antenna phase center through the coordinate descent method, does not need to rely on external sensors, effectively improves the imaging definition, and thus realizes a compact, light and independent handheld high-resolution millimeter wave synthetic aperture radar system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of radar imaging, and particularly relates to a handheld high-resolution millimeter-wave synthetic aperture radar imaging method. BACKGROUND

[0002] Handheld high-resolution millimeter-wave synthetic aperture radar (HR-mmW-SAR) technology represents a major breakthrough in the new generation of portable imaging systems. This technology can achieve penetrating high-resolution imaging in complex environments and adverse weather conditions, and has unique advantages in search and rescue, security checks, and infrastructure assessment. Compared with traditional optical and infrared imaging technologies, HR-mmW-SAR has the following significant features: 1) all-weather working capability; 2) penetrating non-metallic obstacles imaging; 3) compact design for easy handheld operation.

[0003] The back-projection (BP) algorithm is an ideal choice for handheld SAR systems due to its adaptability to arbitrary motion trajectories. However, the practical application of this technology faces two major challenges: first, handheld operation inevitably introduces platform position errors, leading to a significant decline in imaging quality; second, existing compensation methods have obvious limitations. Inertial measurement unit-based solutions, while having high accuracy, increase system complexity and cost, and are difficult to capture hand tremors. Traditional autofocusing algorithms (such as phase gradient autofocus) rely on Fourier transform relationships, and the coherent integration characteristics of the BP algorithm make it impossible to directly apply these methods. This technical bottleneck severely restricts the performance of HR-mmW-SAR in imaging applications.

[0004] With the rapid development of millimeter-wave hardware and signal processing technology, developing new autofocusing methods suitable for the BP algorithm framework has become a current research hotspot. The following difficulties still exist: 1) reliance on external sensors; 2) difficulty in adapting to the special mathematical properties of the BP algorithm; 3) insufficient computational efficiency. Therefore, how to overcome the above difficulties has become a problem to be solved in the field. SUMMARY

[0005] The purpose of the present application is to provide a handheld high-resolution millimeter-wave synthetic aperture radar imaging method (minimum-entropy autofocus back-projection, MEA-BP). This method solves the problem of imaging blur caused by radar position errors introduced by hand jitter.

[0006] Technical solution: A handheld high-resolution millimeter wave synthetic aperture radar imaging method, comprising the following steps:

[0007] Step 1, collect radar signals through a handheld high-resolution millimeter wave radar system, and mix the radar signals to obtain intermediate frequency signals;

[0008] Step 2, perform fast Fourier transform processing and residual video phase removal on the intermediate frequency signals to obtain processed signals;

[0009] Step 3, based on the processed signals, calculate the frequency response corresponding to the distance between the radar antenna phase center and the pixel point through linear interpolation;

[0010] Step 4, based on the frequency response, according to the backward projection imaging, multiply the azimuth phase compensation, and then perform coherent summation to obtain an image expression;

[0011] Step 5, based on the image expression, according to the minimum entropy algorithm, update the position error by calculating the image entropy and selecting the minimum error to obtain an optimal error estimation vector;

[0012] Step 6, correct the motion trajectory of the radar antenna phase center through the optimal error estimation vector, and re-execute the backward projection imaging.

[0013] Further, step 1 is specifically: during radar data acquisition, the handheld high-resolution millimeter wave radar system emits a frequency-modulated continuous wave signal with a duration of along its trajectory, which is reflected by the target and received by the receiver, mixed with the transmitted signal to generate an intermediate frequency signal:

[0014]

[0015] In the formula, is the speed of light, is the carrier wavelength of the signal, is the carrier frequency, is a continuous slow time variable, is a continuous fast time variable, and is the slope of the linear frequency modulation signal, B is the signal bandwidth, R is the instantaneous slant range.

[0016] Further, step 2 is specifically: after fast Fourier transform processing and residual video phase removal, the processed signal is:

[0017]

[0018] Further, step 3 is specifically: calculate the distance between the radar antenna phase center and the pixel point through linear interpolation The corresponding frequency response is:

[0019]

[0020] wherein, is the discrete distance, and respectively represent and range compressed signals at and represent the continuous index of adjacent discrete distance points at .

[0021] Further, step 4 is specifically: according to the back-projection imaging, multiplying the azimuth phase compensation , and then performing coherent summation to obtain an image expression:

[0022]

[0023] wherein, represents the slow time is sampled as points, is the horizontal coordinate of a pixel point, is the vertical coordinate of a pixel point, I represents the coherent imaging intensity value of a specific pixel point in a target scene image, which is a basic unit constituting a final radar image.

[0024] Further, in step 5, the minimum entropy algorithm is specifically:

[0025]

[0026] wherein, represents the information entropy of a radar imaging result, is a pixel value after image normalization;

[0027] The coordinate descent method is used for iterative optimization, only one parameter is updated each time, and other parameters remain unchanged, and the iterative process is recorded as .

[0028] Further, in step 5, the position error is updated by calculating the image entropy and selecting the minimum error to obtain an optimal error estimation vector, which specifically includes the following steps:

[0029] Step 5.1, the position error is initialized as , that is, it is assumed that the error of all trajectory points is zero at the beginning;

[0030] Step 5.2, in the th iteration, the current position error vector is For the nth track point, the update amount is introduced The updated error vector is represented as:

[0031] ;

[0032] In the formula, represents the position error of the mth track point on the radar motion trajectory in the n th iteration process; j N

[0033] Step 5.3, calculate the corresponding image entropy of the updated , and find the that minimizes the image entropy; the optimization problem is described as:

[0034]

[0035] Step 5.4, repeat steps 5.2 and 5.3 for all track points, update each element in the error vector one by one, until the update of the entire track is completed, through multiple iterations, gradually optimize the position error, and finally obtain the optimal error estimation vector .

[0036] The application also discloses a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to realize the steps of the method of the application.

[0037] The application also discloses a computer readable storage medium, which stores a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the method of the application.

[0038] The application also discloses a computer program product, which comprises a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the method of the application.

[0039] Advantages: compared with the prior art, the application has the following obvious advantages:

[0040] (1) The application uses the minimum entropy self-focusing algorithm to realize motion error compensation only by using echo data, without any external sensor, significantly reduces the hardware cost and volume, and is more suitable for portable application scenarios.

[0041] (2) According to the coherent integration characteristics of the BP algorithm, the position error is optimized point by point by using the coordinate descent method, so that the image entropy is minimized, high-efficiency self-focusing under the BP framework is realized, and the limitations of traditional technologies are broken through.

[0042] ​​(3) The method has higher peak side ratio and integral side lobe ratio compared with the prior art maximum image intensity back projection method and maximum image definition back projection method.

[0043] (4) The application can realize high resolution imaging by combining commercial millimeter wave radar, and provides a feasible solution for low cost portable SAR system in search and rescue, safety inspection and other fields. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 It is a work flow diagram of a handheld high-resolution millimeter wave synthetic aperture radar imaging method in the embodiment of the application;

[0045] Figure 2 It is a work flow diagram of a minimum entropy self-focusing algorithm in the handheld high-resolution millimeter wave synthetic aperture radar imaging method in the embodiment of the application;

[0046] Figure 3 It is a simulation result comparison diagram of the handheld high-resolution millimeter wave synthetic aperture radar imaging method and other methods in the embodiment of the application: (a) random position error, (b) BP without position error, (c) BP with position error, (d) MPA-BP, (e) MSA-BP, (f) MEA-BP;

[0047] Figure 4 It is an azimuth profile and a range profile of a center point under simulation results in the embodiment of the application: (a) azimuth profile, (b) range profile;

[0048] Figure 5 It is an HR-mmW-SAR system and an experimental scene setting used in the embodiment of the application: (a) HR-mmW-SAR system, (b) experimental scene;

[0049] Figure 6 It is an experimental result comparison diagram of the handheld high-resolution millimeter wave synthetic aperture radar imaging method and other methods in the embodiment of the application: (a) BP without automatic focusing, (b) MPA-BP, (c) MSA-BP, (d) MEA-BP;

[0050] Figure 7 It is an azimuth profile and a range profile of target 1 under an experimental scene in the embodiment of the application: (a) azimuth profile, (b) range profile;

[0051] Figure 8 It is an azimuth profile and a range profile of target 2 under an experimental scene in the embodiment of the application: (a) azimuth profile, (b) range profile. DETAILED DESCRIPTION

[0052] The technical solutions of the application are further described below with reference to the drawings. A handheld high-resolution millimeter wave synthetic aperture radar imaging method of the application has a whole working flow as shown in the figure Figure 1 The steps of the handheld high-resolution millimeter wave synthetic aperture radar imaging are as follows:

[0053] Step 1. During radar data collection, the HR-mmW-SAR system transmits a frequency-modulated continuous wave signal with a duration of along its trajectory, which is reflected by the target and received by the receiver, mixed with the transmitted signal to generate an intermediate frequency signal as follows:

[0054]

[0055] In the formula, is the speed of light, is the carrier wavelength of the signal, is the carrier frequency, is a continuous slow time variable, is a continuous fast time variable, and is the slope of the linear frequency modulation signal, B is the signal bandwidth, R is the instantaneous slant range.

[0056] Step 2. After fast Fourier transform processing and residual video phase removal, the baseband signal is as follows:

[0057]

[0058] Step 3. The distance between the radar antenna phase center and the pixel point is calculated by linear interpolation The corresponding frequency response is as follows:

[0059]

[0060] In the formula, is the discrete distance, and represent the range-compressed signals at and respectively.

[0061] Step 4. According to BP imaging, multiply by the azimuth phase compensation Then perform coherent summation to obtain the image expression:

[0062]

[0063] In the formula, represents the slow time is sampled as points, is the horizontal coordinate of the pixel point, is the pixel point ordinate.

[0064] Step 5, according to the minimum entropy algorithm, the position error is updated by calculating the image entropy and selecting the minimum error, and the entropy function of the BP image is:

[0065]

[0066] The present application adopts the coordinate descent method for iterative optimization, and only one parameter is updated each time, and other parameters remain unchanged. The iterative process is recorded as .

[0067] The workflow of step 5 is shown in Figure 2 , and specifically includes:

[0068] Step 5.1, the initial position error is , that is, it is assumed that the error of all trajectory points is zero at the beginning.

[0069] Step 5.2, in the first iteration, the current position error vector is , for the nth trajectory point, by introducing the update amount , the updated error vector is represented as:

[0070]

[0071] Step 5.3, for the updated , the corresponding image entropy is calculated and the that makes the image entropy minimum is found. The optimization problem is described as:

[0072]

[0073] Step 5.4, repeat step 5.2 and step 5.3 for all trajectory points, update each element in the error vector one by one until the update of the entire trajectory is completed. Through multiple iterations, the position error is gradually optimized, and finally the optimal error estimation vector is obtained.

[0074] Step 6, using to correct the motion trajectory of the radar antenna phase center, and re-executing the BP imaging.

[0075] In further embodiments, the main simulation parameters of the present application in the simulation scene are shown in Table 1.

[0076] Table 1

[0077]

[0078] The simulation results are as followsFigure 3 As shown, the generated random position error is as follows Figure 3 As shown in (a), the focusing results of simulation data for BP without positional error, BP with positional error, maximum intensity autofocus back-projection (MPA-BP), maximum sharpness autofocus back-projection (MSA-BP), and MEA-BP are as follows: Figure 3 As shown in (b)-(f) of the simulation results. The image entropy, sharpness, and intensity indices of the simulation results are shown in Table 2.

[0079] Table 2

[0080]

[0081] Figure 4 yes Figure 3 (b) A cross-sectional view of the center point in the middle circle. Figure 4 (a) in the diagram is an azimuth profile. Figure 4 (b) in the table is the distance profile. The dashed line represents the BP result without positional error, the dotted line represents the BP result with positional error, the dotted-dash line represents the MPA-BP result, the thin solid line represents the MSA-BP result, and the thick solid line represents the result of the method of this invention. The PSLR and ISLR of the center point are shown in Table 3, where the azimuth profile and the distance profile are distinguished by the prefixes A and R, respectively.

[0082] Table 3

[0083]

[0084] In a further embodiment, the present invention performs imaging in an experimental scenario, using an HR-mmW-SAR system and target settings in the scenario as follows. Figure 5 As shown, Figure 5 (a) in the figure represents the HR-mmW-SAR system. Figure 5 (b) in the table shows the target settings in the experimental scenario. The radar parameter settings are shown in Table 1.

[0085] Figure 6 This is a comparison chart of experimental results between the method of this invention and other methods. Figure 6 In Table 4, (a)-(d) show the results of BP without autofocus, MPA-BP, MSA-BP, and MEA-BP using the method of this invention, respectively. The image entropy, sharpness, and intensity indices of the experimental results are shown in Table 4.

[0086] Table 4

[0087]

[0088] Figure 7 for Figure 5 the azimuth profile and the range profile of target 1 in (b) in FIG. 5, Figure 7 (a) in FIG. 5 is an azimuth profile, Figure 7 (b) in FIG. 5 is a range profile. Figure 8 for Figure 5 the azimuth profile and the range profile of target 2 in (b) in FIG. 6, Figure 8 (a) in FIG. 6 is an azimuth profile, Figure 8 (b) in FIG. 6 is a range profile. The dotted line is the BP result without auto-focusing, the dot line is the MPA-BP result, the dot-dash line is the MSA-BP result, and the solid line is the MEA-BP result. The PSLR and ISLR of the two targets are shown in Table 5.

[0089] From the comparison of the results of the method of the present application and other methods in the simulation scene and the experimental scene, the imaging target point feature is clearly visible, and at the same time has the lowest image entropy and higher definition. In terms of focusing indicators, PSLR and ISLR perform excellently. In summary, the present application can effectively adapt to target position error, suppress defocus and noise interference, and realize high-quality motion compensation.

[0090] The above-described specific embodiments further detail the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are merely specific embodiments of the present application and are not intended to limit the scope of the present application. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principles of the present application shall fall within the scope of the present application.

[0091] Table 5

[0092]

Claims

1. A handheld high-resolution millimeter-wave synthetic aperture radar imaging method, characterized in that, Includes the following steps: Step 1: Acquire radar signals using a handheld high-resolution millimeter-wave radar system, and mix the radar signals to obtain an intermediate frequency signal; Step 2: Perform Fast Fourier Transform and residual video phase removal on the intermediate frequency signal to obtain the processed signal; Step 3: Based on the processed signal, calculate the frequency response corresponding to the distance between the radar antenna phase center and the pixel point through linear interpolation; Step 4: Based on the back projection imaging, multiply the frequency response by the azimuth phase compensation and then perform coherent summation to obtain the image expression; Step 5: Based on the image expression, according to the minimum entropy algorithm, the position error is updated by calculating the image entropy and selecting the minimum error to obtain the optimal error estimation vector; In step 5, the minimum entropy algorithm is specifically as follows: ； In the formula, The information entropy representing the radar imaging results, These are the pixel values ​​after image normalization. The x-coordinate of the pixel is The ordinate of the pixel; Iterative optimization is performed using the coordinate descent method, updating only one parameter at a time while keeping other parameters unchanged. The iterative process is denoted as... ; In step 5, the step of calculating the image entropy and selecting the minimum error to update the position error to obtain the optimal error estimation vector specifically includes the following steps: Step 5.1: Initialize the position error as follows: That is, assume that the error of all trajectory points is zero at the beginning; Step 5.2, in the... In this iteration, the current position error vector is For the first Each trajectory point, by introducing update values The updated error vector is represented as: ； In the formula, Indicates the first In the nth iteration process, the radar trajectory on the nth... Position error of each trajectory point; Step 5.3, update the... Calculate the corresponding image entropy And find the one that minimizes image entropy. The optimization problem is described as follows: ； Step 5.4: Repeat steps 5.2 and 5.3 for all trajectory points, updating each element in the error vector one by one until the entire trajectory is updated. Through multiple iterations, the position error is gradually optimized, and finally the optimal error estimation vector is obtained. ; Step 6: Correct the motion trajectory of the radar antenna phase center using the optimal error estimation vector, and re-execute the back projection imaging.

2. The handheld high-resolution millimeter-wave synthetic aperture radar imaging method according to claim 1, characterized in that, Step 1 specifically involves: During radar data acquisition, the handheld high-resolution millimeter-wave radar system transmits along its trajectory for a duration of [duration missing]. The frequency-modulated continuous wave signal, after being reflected by the target, is received by the receiver and mixed with the transmitted signal to generate an intermediate frequency signal: ； In the formula, At the speed of light, The carrier wavelength of the signal. For carrier frequency, For continuous slow-time variables, For continuous fast-time variables, and It is the slope of the linear frequency modulated signal. For signal bandwidth, This represents the instantaneous slant distance.

3. The handheld high-resolution millimeter-wave synthetic aperture radar imaging method according to claim 2, characterized in that, Step 2 specifically involves performing a Fast Fourier Transform (FFT) on the intermediate frequency (IF) signal and removing lingering video phase, resulting in the following processed signal: 。 4. The handheld high-resolution millimeter-wave synthetic aperture radar imaging method according to claim 1, characterized in that, Step 3 specifically involves calculating the distance between the radar antenna phase center and the pixel using linear interpolation. The corresponding frequency response is: ； In the formula, For discrete distance, and They represent and Local range compressed signal, and Indicates distance A continuous index of adjacent discrete distance points.

5. The handheld high-resolution millimeter-wave synthetic aperture radar imaging method according to claim 4, characterized in that, Step 4 specifically involves: Based on the back projection imaging, multiplied by the azimuth phase compensation Then, perform coherent summation to obtain the image expression: ； In the formula, Indicates slow time Sampling as I represents a specific pixel in the target scene image. The coherent imaging intensity value is the basic unit that constitutes the final radar image.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method of claim 1.

7. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.

Images