High-resolution one-dimensional imaging method based on multi-scattering point model

Abstract

The present invention discloses a high-resolution one-dimensional imaging method based on a multi-scattering point model with high imaging accuracy. The method comprises: (10) radar parameter acquisition: acquiring radar parameters, including pulse train length, step initial frequency, pulse repetition period, pulse width, frequency step interval, and initial phase; (20) multi-scattering point target model establishment: obtaining a target surface element model based on a three-dimensional target geometric model, calculating the radar cross-sectional area of ​​the visible surface element in each distance unit, and performing vector synthesis according to the distance unit to obtain the radar cross-sectional area and spatial coordinates of each scattering point in different distance units, changing the intersection condition, calculating the scattering point parameters in each distance unit under different intersection conditions, and performing statistical processing on the scattering point parameters to obtain a multi-scattering point model of the target; (30) high-resolution one-dimensional imaging: combining the multi-scattering point target model and the radar parameters to obtain a one-dimensional range image of the target.

Description

Technical Field

[0001] The invention belongs to the technical field of imaging fuses, and in particular relates to a high-resolution one-dimensional imaging method based on a multi-scattering point model. Background Art

[0002] Radar high-resolution range image refers to the use of broadband radar signals to obtain fuze target information. Its characteristic is that a high-frequency signal of a certain wavelength is emitted, and the received echo signal is subjected to Fourier transform to obtain a high-resolution range image. The radar high-resolution range image well reflects the radial distance distribution details and structural information of the target along the radar line of sight. The slow variation of the scattering point position in the range image with the change of the target posture can serve as the basis for target recognition research.

[0003] For most high-resolution radars, actual target verification requires multiple verifications due to the high speed of target motion and subsequent attitude changes. This is costly and wastes resources. Modeling actual targets allows for rapid and multiple design and verification. Traditional multi-scattering point target simulation methods based on multiple standard bodies simplify computations and reduce the amount of data required for inverse Fourier transforms compared to the original method of synthesizing all visible elements, but at the expense of accuracy.

[0004] Therefore, the problems existing in the existing technology are: the radar high-resolution one-dimensional imaging has a large amount of computation and the accuracy is not high enough. Summary of the Invention

[0005] The object of the present invention is to provide a high-resolution one-dimensional imaging method based on a multi-scattering point model with high imaging accuracy.

[0006] The technical solution for achieving the purpose of the present invention is: a method for establishing a high-resolution one-dimensional imaging fuze target model, comprising the following steps:

[0007] (10) Radar parameter acquisition: Acquire radar parameters, including pulse train length, step initial frequency, pulse repetition period, pulse width, frequency step interval, and initial phase;

[0008] (20) Establishment of multi-scattering point target model: Based on the target three-dimensional geometric model, the target surface element model is obtained. The radar cross-sectional area of ​​the visible surface element in each distance unit is calculated, and vector synthesis is performed according to the distance unit to obtain the radar cross-sectional area and spatial coordinates of each scattering point in different distance units. The intersection conditions are changed, and the scattering point parameters in each distance unit under different intersection conditions are calculated. The scattering point parameters are statistically processed to obtain the multi-scattering point model of the target.

[0009] (30) High-resolution one-dimensional imaging: Combining the multi-scattering point target model and radar parameters, a one-dimensional range image of the target is obtained.

[0010] Compared with the prior art, the present invention has the following significant advantages:

[0011] 1. High imaging accuracy: The present invention adopts a modeling method based on a multi-scattering point model, which makes the imaging accuracy high and solves the problem of insufficient imaging accuracy in the prior art.

[0012] 2. Small amount of computation: The present invention adopts multi-scattering point modeling, which reduces the amount of computation of the surface element model and solves the problem of large amount of computation in the prior art.

[0013] The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is the main flow chart of the high-resolution one-dimensional imaging based on the multi-scattering point model of the present invention.

[0015] Figure 2 for Figure 1 Flowchart of the steps for establishing a multi-scattering point target model

[0016] Figure 3 for Figure 2 Flowchart of the steps for extracting the target surface element model.

[0017] Figure 4 for Figure 3 Flowchart of the steps for determining whether a facet is visible.

[0018] Figure 5 for Figure 2 Flowchart of the steps for calculating scattering point parameters. DETAILED DESCRIPTION

[0019] like Figure 1 As shown, the high-resolution one-dimensional imaging method based on the multi-scattering point model of the present invention includes the following steps:

[0020] (10) Radar parameter acquisition: Acquire radar parameters, including pulse train length, step initial frequency, pulse repetition period, pulse width, frequency step interval, and initial phase;

[0021] (20) Establishment of multi-scattering point target model: Based on the target three-dimensional geometric model, the target surface element model is obtained. The radar cross-sectional area of ​​the visible surface element in each distance unit is calculated, and vector synthesis is performed according to the distance unit to obtain the radar cross-sectional area and spatial coordinates of each scattering point in different distance units. The intersection conditions are changed, and the scattering point parameters in each distance unit under different intersection conditions are calculated. The scattering point parameters are statistically processed to obtain the multi-scattering point model of the target.

[0022] like Figure 2As shown, the (20) multi-scattering point target model establishment step includes:

[0023] (21) Extracting the target facet model: According to the target three-dimensional geometric model, the vertex coordinates and external normal vectors of each target triangular facet in the target coordinate system are obtained, and the visible surface is judged;

[0024] like Figure 3 As shown, the step (21) of extracting the facet model includes:

[0025] (211) Constructing a block-based target face model: The target face model consists of a block structure. A single block structure contains a data header and a data body. The data header contains the meaning and length of the block, and the data body contains the specific data content. The data content is based on the three-dimensional geometric model of the target, and stores the vertex coordinates (x mt ,y mt , z mt )(m=1,2,3) and the external normal vector (x n ,y n , z n ).

[0026] (212) Determine whether the surface element is visible: read the external normal and incident wave direction vector of the target surface element in the target coordinate system and the receiving direction vector For any triangle surface S, calculate the external normal and the incident wave direction vector The inner product of and receiving direction vector The inner product of . When , the surface element S can be illuminated by the incident wave. When , the surface element S can be detected by the receiver. and The surface element is the visible surface element in the detection area. The specific process is as follows Figure 4 shown.

[0027] (22) Dividing the surface element according to the distance unit: Dividing the target surface element model into surface elements according to the distance unit;

[0028] Furthermore, the step (22) of dividing the surface into elements according to the distance units includes:

[0029] (221) Range unit setting: dr is the range resolution of the stepped frequency radar signal, which can also be called a range resolution unit. Its expression is:

[0030]

[0031] Where: B is the signal bandwidth, C is the speed of light, N is the pulse train length, and Δf is the frequency step interval. The target is divided into M range units through the setting of range units, and each range unit contains all the visible elements in the range unit.

[0032] (222) Distance intersection setting: When a visible surface element is between two distance units, the distance unit where the center point of the visible surface element is located is selected as the selected distance unit. If the center point is located on the intersection line, then it is determined that two of the three vertices in the visible surface element are in the same distance unit as the selected distance unit.

[0033] (23) Scattering point parameter calculation: Calculate the radar cross-section of the visible element in each distance unit and perform vector synthesis based on the distance unit to obtain the radar cross-section and spatial coordinates of each scattering point in different distance units;

[0034] like Figure 5 As shown, the (23) scattering point parameter calculation step includes:

[0035] (231) Calculation of visible element radar cross-sectional area: The induced current on the surface of any object is approximated by physical optics. Once the induced current on the target surface is known, the far-field scattering field intensity of the target can be calculated. The limiting conditions of the far-field scattering field are as follows:

[0036]

[0037] Where R is the distance between the target and the fuze, λ is the wavelength of the transmitted signal, and D is the maximum linear dimension of the target. When the target is a bin, its maximum linear dimension is determined by the longest side length. Therefore, when the bin is sufficiently small, the far-field condition can be approximately met within the near-field region. The radar cross section (RCS) of a single bin is calculated as:

[0038]

[0039] Where:

[0040]

[0041] T=|ω r |cosθ

[0042]

[0043] In the above formulas, λ is the wavelength of the transmitted signal, A is the area of ​​the triangular element, is the surface element's external normal vector, is the electric field polarization direction, is the magnetic field polarization direction, is the incident direction unit vector, is the scattering direction unit vector, θ is ω r The angle between the triangle surface projection vector and the triangle surface projection vector.

[0044] (232) Radar cross-section synthesis: Let the total number of visible pixels be K, and the distance between each pixel and the center of the fuze be d. i , RCS is Each visible element is synthesized into M scattering points according to the distance unit. The RCS (in dBm) of a scattering point within each distance unit of the target is 2 The calculation method is as follows:

[0045]

[0046] Where: σ l M is the RCS size of the scattered point after the synthesis of the l-th range unit, l The number of visible facets contained in the lth distance unit; is the RCS of the ith bin, d i is the distance from the ith element to the fuse;

[0047] (233) Space coordinate synthesis: The center coordinates (x iu ,y iu , z iu )(i=1,2,3...K), the formula is as follows:

[0048]

[0049]

[0050] The spatial coordinate position information of the surface element in each range unit is synthesized according to the radar cross-sectional area, and the spatial position coordinates (x lv ,y lv , z lv )(l=1,2,3...M); the formula is as follows:

[0051]

[0052] (24) Calculation of multi-scattering point model parameters: Change the intersection conditions, calculate the scattering point parameters in each distance unit under different intersection conditions, and perform statistical processing on the scattering point parameters to obtain the multi-scattering point model of the target.

[0053] Furthermore, the (24) multiple measurement statistical step includes:

[0054] (241) Changing the rendezvous conditions: Under different rendezvous conditions, the scattering point parameters under different rendezvous conditions are obtained, including the synthetic radar scattering cross-section and spatial coordinates.

[0055] (242) Data statistics: Statistical processing is performed on the obtained scattering point parameters, and the mean and variance are calculated. The results are statistically analyzed to verify that they conform to the normal distribution, and the multi-scattering point model of the target is obtained. The final data format is as follows Table 1:

[0056]

[0057] (30) High-resolution one-dimensional imaging: Combining the multi-scattering point target model and radar parameters, a one-dimensional range image of the target is obtained.

[0058] Furthermore, the (30) high-resolution one-dimensional imaging step is specifically as follows:

[0059] According to different intersection conditions, the corresponding multi-scattering point model is selected. Combined with the corresponding radar parameters, the multi-scattering point model data is vector synthesized to obtain the stepped frequency echo signal of the complete target. The stepped frequency echo signal is coherently demodulated to obtain the echo mixing signal. Finally, the echo mixing signal is inverse Fourier transformed to obtain the one-dimensional range image.

[0060] The above process demonstrates that the method of the present invention constructs a 3D geometric model of the target, followed by a surface element model, and finally a multi-scattering point model. This multi-scattering point target model simplifies implementation and increases the reliability of the computational results when designing and validating the imaging algorithm for a high-resolution 1D imaging fuze.

Claims

1. A high-resolution one-dimensional imaging method based on a multi-scattering point model, characterized in that: The steps include: (10) Radar parameter acquisition: Acquire radar parameters, including pulse train length, step initial frequency, pulse repetition period, pulse width, frequency step interval, and initial phase; (20) Establishment of multi-scattering point target model: Based on the target three-dimensional geometric model, the target surface element model is obtained. The radar cross-sectional area of ​​the visible surface element in each distance unit is calculated, and vector synthesis is performed according to the distance unit to obtain the radar cross-sectional area and spatial coordinates of each scattering point in different distance units. The intersection conditions are changed, and the scattering point parameters in each distance unit under different intersection conditions are calculated. The scattering point parameters are statistically processed to obtain the multi-scattering point model of the target; (21) Extracting the target facet model: According to the target three-dimensional geometric model, the vertex coordinates and external normal vectors of each target triangular facet in the target coordinate system are obtained, and the visible surface is judged; (211) Constructing a block-based target face model: The target face model consists of a block structure. A single block structure contains a data header and a data body. The data header contains the meaning and length of the block, and the data body contains the specific data content. The data content is based on the three-dimensional geometric model of the target, and stores the vertex coordinates (x mt ,y mt , z mt )(m=1,2,3) and the external normal vector (x n ,y n , z n ); (212) Determine whether the surface element is visible: read the external normal and incident wave direction vector of the target surface element in the target coordinate system and the receiving direction vector For any triangle surface S, calculate the external normal and the incident wave direction vector The inner product of and receiving direction vector The inner product of When , the surface element S can be illuminated by the incident wave. When , the surface element S can be detected by the receiver; only when and The surface element is the visible surface element in the detection area; (22) Dividing the surface element according to the distance unit: Dividing the target surface element model into surface elements according to the distance unit; (221) Range unit setting: dr is the range resolution of the stepped frequency radar signal, which can also be called a range resolution unit. Its expression is: Where: B is the signal bandwidth, C is the speed of light, N is the pulse train length, and Δf is the frequency step interval. The target is divided into M range units through the setting of range units, and each range unit contains all the visible elements in the range unit. (222) Distance intersection setting: When a visible surface element is located between two distance units, the distance unit where the center point of the visible surface element is located is selected as the selected distance unit. If the center point is located on the intersection line, two of the three vertices in the visible surface element are located in the same distance unit as the selected distance unit; (23) Scattering point parameter calculation: Calculate the radar cross-section of the visible element in each distance unit and perform vector synthesis based on the distance unit to obtain the radar cross-section and spatial coordinates of each scattering point in different distance units; (231) Calculation of visible element radar cross-sectional area: The induced current on the surface of any object is approximated by physical optics. Once the induced current on the target surface is known, the far-field scattering field intensity of the target can be calculated. The limiting conditions of the far-field scattering field are as follows: Where R is the distance between the target and the fuze, λ is the wavelength of the transmitted signal, and D is the maximum linear dimension of the target. The radar cross section (RCS) of a single bin is calculated as follows: Where: T=|ω r |cosθ In the above formulas, λ is the wavelength of the transmitted signal, A is the area of ​​the triangular element, is the surface element's external normal vector, is the electric field polarization direction, is the magnetic field polarization direction, is the incident direction unit vector, is the scattering direction unit vector, θ is ω r The angle between it and the projection vector of the triangle surface element; (232) Radar cross-section synthesis: Let the total number of visible pixels be K, and the distance between each pixel and the center of the fuze be d. i , RCS is Each visible element is synthesized into M scattering points according to the distance unit. The RCS (in dBm) of a scattering point within each distance unit of the target is 2 The calculation method is as follows: Where: σ l M is the RCS size of the scattered point after the synthesis of the l-th range unit, l The number of visible facets contained in the lth distance unit; is the RCS of the ith bin, d i is the distance from the ith element to the fuse; (233) Space coordinate synthesis: The center coordinates (x iu ,y iu , z iu )(i=1,2,3...K), the formula is as follows: The spatial coordinate position information of the surface element in each range unit is synthesized according to the radar cross-sectional area, and the spatial position coordinates (x lv ,y lv , Z lv )(l=1,2,3...M); the formula is as follows: (24) Calculation of multi-scattering point model parameters: changing the intersection conditions, calculating the scattering point parameters in each distance unit under different intersection conditions, and performing statistical processing on the scattering point parameters to obtain the multi-scattering point model of the target; (241) Changing the intersection conditions: Under different intersection conditions, the scattering point parameters under different intersection conditions are obtained, including the synthetic radar cross-sectional area and spatial coordinates; (242) Data statistics: Statistical processing is performed on the obtained scattering point parameters to calculate the mean and variance. The results are statistically analyzed to verify that they conform to the normal distribution and obtain the target multi-scattering point model; (30) High-resolution one-dimensional imaging: Combining the multi-scattering point target model and radar parameters, a one-dimensional range image of the target is obtained.

2. The high-resolution one-dimensional imaging method according to claim 1, characterized in that: The (30) high-resolution one-dimensional imaging step is specifically as follows: According to different intersection conditions, the corresponding multi-scattering point model is selected. Combined with the corresponding radar parameters, the multi-scattering point model data is vector synthesized to obtain the stepped frequency echo signal of the complete target. The stepped frequency echo signal is coherently demodulated to obtain the echo mixing signal. Finally, the echo mixing signal is inverse Fourier transformed to obtain the one-dimensional range image.

Images