A radar target attribute scattering center extraction method combining calculation and measurement data

Abstract

A radar target attribute scattering center extraction method combining calculation and measurement data, obtains one-dimensional single-frequency scan angle simulation data and two-dimensional scan frequency scan angle test data, selects one-dimensional single-frequency scan angle simulation data to calculate time-frequency image, detects local scattering center according to Doppler frequency characteristics, carries out IRT transformation on the time-frequency image to obtain position information and amplitude information of the local scattering center LSC, carries out one-dimensional peak value detection on each row of the time-frequency image, sets a detection threshold and minimum DSC length of distributed scattering center DSC, detects and extracts DSC information, solves the accurate position of the DSC, substitutes the obtained scattering center information into a scattering center model to obtain a reconstructed scattering field, and adjusts the relationship between the time-frequency image amplitude and the electric field amplitude to make the model amplitude more accurate. The application reduces the data amount required for attribute scattering center parameter estimation, improves the operation efficiency while ensuring the accuracy of the model, and can effectively improve the accuracy of the reconstructed model RCS.

Description

Technical Field

[0001] This invention relates to the field of radar target feature extraction technology, and in particular to a method for extracting the scattering center of radar target attributes. Background Technology

[0002] In the research methods of radar target characteristic analysis, the scattering center model is an important tool for describing the electromagnetic scattering characteristics of radar targets. With the continuous deepening of theoretical research and the continuous improvement of radar signal processing technology, the estimation of attribute scattering center (ASC) parameters has become an important research topic in the field of target identification. By analyzing and processing target scattering data, the structural characteristics of radar targets can be analyzed more intuitively, such as the physical size of the target body and the structure of typical components. Furthermore, the near-field echo of the target can be extracted and analyzed through scattering center information, thereby quickly identifying unknown radar targets. In the application of target characteristics, extracting the scattering center from the test data of scaled-down targets can provide an important reference for the study of non-cooperative targets. Since the attitude of the target tested in the anechoic chamber is limited, it is impossible to obtain scattering data for all attitude angles. This invention combines one-dimensional simulation data with two-dimensional test data, which can not only obtain scattering center data for all attitudes, but also improve the accuracy of the scattering center model through the combination of calculation and measurement. Summary of the Invention

[0003] The purpose of this invention is to provide a radar target attribute scattering center extraction method that combines computational and measurement data, which reduces the amount of data required for attribute scattering center parameter estimation, improves computational efficiency while ensuring model accuracy, and can effectively improve the accuracy of the reconstructed model's RCS.

[0004] To achieve the above objectives, the present invention provides a method for extracting the scattering center of radar target attributes, comprising the following steps:

[0005] Step S1: Obtain the one-dimensional single-frequency sweep angle simulation data and two-dimensional sweep frequency sweep angle test data of the target. Based on the far-field radar scattering cross section and phase, the far-field scattering data of the target is uniformly represented as a complex electric field.

[0006] Step S2: Select one-dimensional single-frequency sweep angle simulation data to calculate the time-frequency image;

[0007] Step S3: Based on the Doppler frequency characteristics, detect the local scattering center, perform IRT transformation on the time-frequency image, and obtain the location and amplitude information of the local scattering center LSC;

[0008] Step S4: Based on the Doppler frequency characteristics, one-dimensional peak detection is performed on each row of the time-frequency image. The detection threshold and minimum DSC length of the distributed scattering center DSC are set, and DSC information is detected and extracted.

[0009] Step S5: Determine the precise position of the DSC based on the appearance angle and length of the DSC obtained in step S4.

[0010] Step S6: Substitute the obtained scattering center information into the scattering center model to obtain the reconstructed scattering field. Then, adjust the model according to the relationship between the time-frequency image amplitude and the electric field amplitude to make the amplitudes of the local scattering center LSC and the distributed scattering center DSC models more accurate.

[0011] The target far-field scattering data is uniformly represented as a complex electric field. The correspondence between the far-field scattering results and the electric field in the simulation software and testing system is as follows:

[0012] E=10 (σ / 20) ×exp(-1j×φ×π / 180)

[0013] In the formula, σ is the radar cross section (RCS) in dBsm, φ is the phase in °, j is a complex unit, and E is the complex electric field.

[0014] Step S2 includes the following steps:

[0015] S2.1. Select sweep angle data at a single frequency from the one-dimensional simulation data and the two-dimensional test data respectively as the initial input. Determine the window size according to the number of sweep angles, and use smooth pseudo-Wigner-Ville distribution SPWVD for time-frequency transformation:

[0016]

[0017] In the formula, windows g(u) and h(τ) are both real even functions, and s(t) is the one-dimensional complex data defined in step S1;

[0018] After time-frequency transformation, the time-frequency characteristics of the scattered electric field correspond to the variation characteristics of the Doppler frequency with time.

[0019] Step S2.2: After obtaining the time-frequency image, first calculate the overall mean of the image. Set the image edge threshold to 1 / 20 of the overall mean. Compare the mean values ​​of the upper and lower sides from left to right to obtain the positions of the upper and lower edges.

[0020] The radar target attribute scattering center extraction method as described in claim 3 is characterized in that, in step S2.2, the edge position is checked to see if there is oversampling or undersampling. The edge being located between one-half and three-quarters of the boundary is normal, the edge being below one-half of the boundary is undersampling, and the edge being above three-quarters is oversampling. If there is undersampling or oversampling, the input data needs to be reconstructed by interpolation or decimation methods, and step S2.1 is repeated.

[0021] Step S3 includes the following steps:

[0022] Step S3.1, the expression for the Local Scattering Center (LSC) model is:

[0023]

[0024] In the formula, f is the radar frequency, φ is the visible azimuth angle range of the scattering center, and r i The scattering center position vector, A is the radar's viewpoint direction vector. i The amplitude at the scattering center;

[0025] The Doppler frequency f at the LSC scattering center Di (t) is represented as:

[0026]

[0027] In the formula, ζ represents the radar perspective. With the scattering center position r i The angle between them f is the distance of the scattering center from the radar viewpoint. Di It exhibits a sinusoidal variation as ζ changes;

[0028] Step S3.2: Process the time-frequency graph using the Radon Inverse Transform (IRT):

[0029]

[0030] In the formula, φ is the radar azimuth angle, and k x =vcosφ,k y =vsinφ, for a single sine curve The phase of the sine curve is represented by the time-frequency graph after IRT. It is represented by a fixed point, with the location being... Extracting the corresponding coordinates (x0, y0), the amplitude and phase of the sine curve corresponding to the LSC are expressed as follows:

[0031]

[0032]

[0033] Step S3.3: Obtain all LSC positions of the simulation data and test data respectively. Select the distance difference according to the target to be tested, extract the intersection of LSC positions as the final LSC position, use the average of the two values ​​for amplitude, and use the maximum visible range of the two values ​​for visible range.

[0034] Step S4 includes the following steps:

[0035] Step S4.1, the model expression for the distributed scattering center DSC is:

[0036]

[0037] In the formula, f is the radar frequency, and φ i L represents the visible angle position of DSC. i r is the DSC length parameter. i The scattering center position vector, A is the radar's viewpoint direction vector. i The amplitude of the scattering center is represented by a vertical bright line in the time-frequency graph;

[0038] Step S4.2: On the same column, detect the maximum value information by row. If the extreme value is higher than the set DSC threshold, and the extreme value extracted after the data in the column is close to the extreme value before the column, or the vertical bright line visibility angle of the adjacent column is close, it is considered to be the same feature and merged. Set the minimum DSC length to 2% of the number of data points. Line features with a length lower than this are not considered.

[0039] The difference in Doppler frequencies between the two endpoints of the DSC is:

[0040]

[0041] In the formula, A and B are the two endpoints of the vertical bright line, x and y are the corresponding coordinates, and f D The Doppler frequency is the DSC frequency at the visible angle. and Perpendicular, the line segment is obtained based on the difference in Doppler frequencies. The actual length L i r Combining the DSC model and the time-frequency transformation formula, the difference in Doppler frequencies Δf D With the model's true length L i r and DSC length parameter L i The relationship between them is:

[0042]

[0043] In the formula, L i r L represents the actual length of the model. i For DSC length parameter, For wave vectors;

[0044] Step S4.3: Obtain all DSC angles from the simulation data and the test data respectively, set the angle difference to 0.5°, extract the intersection of the DSC occurrence angles as the final DSC occurrence angle, and use the average of the two for the length.

[0045] In step S5, based on the angle at which the DSC appears in step S4, two-dimensional sweep frequency test data at the corresponding angle are selected to form a one-dimensional distance image. All peak points of the one-dimensional distance image are found. According to the order of peak amplitude, the DSC position is compared to see if it matches the Doppler frequency range in the time-frequency image. If the distance matches the corresponding DSC length, the corresponding one-dimensional distance image position is determined to be the two endpoints of the DSC.

[0046] In step S6, the RCS is generated based on the scattering center model LSC and compared with the input single-frequency sweep RCS. The model accuracy is judged by the variance of the amplitude difference array between the two. Since the time-frequency transformation will affect the electric field amplitude, it is necessary to calculate the ratio of the peak value of the input RCS to the peak value of the ASC reconstructed RCS, and use it as an adjustment coefficient to multiply the amplitude parameter of the ASC model.

[0047] This invention employs a combined computational and measurement method for extracting attribute scattering centers and estimating parameters, which is of great value for the study and analysis of radar target characteristics. Compared with HRRP history maps, SAR imaging, and ISAR imaging, this algorithm reduces the amount of data required for attribute scattering center parameter estimation, improves computational efficiency while ensuring model accuracy, and can effectively improve the accuracy of the reconstructed model's RCS. Attached Figure Description

[0048] Figure 1 This is a flowchart of the present invention.

[0049] Figure 2 This is a comparison chart of the RCS of the target simulation data and the scattering center model reconstruction.

[0050] Figure 3 It is the time-frequency plot of the RCS of the target simulation data. Detailed Implementation

[0051] The following is based on Figures 1-3 The preferred embodiments of the present invention will be described in detail below.

[0052] The scattering center can be obtained by processing simulation data and test data. This algorithm requires one-dimensional single-frequency sweep angle data and two-dimensional sweep frequency sweep angle data. Since the simulation algorithm is relatively slower than the test algorithm and the confidence level of the simulation algorithm is relatively lower than that of the test data, in order to improve the speed of constructing the scattering center model, this invention adopts a method of combining simulation data and test data to extract the scattering center.

[0053] To ensure the integrity of the scattering center model's all-around information, single-frequency sweep data from the entire field of view was used as input for the time-frequency image, and combined with two-dimensional test data, the scattering center parameters were further optimized.

[0054] like Figure 1As shown, the present invention provides a method for extracting the scattering center of radar target attributes, comprising the following steps:

[0055] Step S1: Obtain the one-dimensional single-frequency sweep angle simulation data and two-dimensional sweep frequency sweep angle test data of the target. Based on the far-field RCS (radar cross section) and phase, the far-field scattering data of the target is uniformly represented as a complex electric field.

[0056] The target far-field scattering data is uniformly represented as a complex electric field. The far-field scattering results of simulation software and testing systems are generally expressed as RCS and phase, and their correspondence with the electric field is as follows:

[0057] E=10 (σ / 20) ×exp(-1j×φ×π / 180)

[0058] In the formula, σ is the RCS in dBsm, φ is the phase in °, j is the complex unit, and E is the complex electric field.

[0059] Step S2: Select one-dimensional single-frequency sweep angle simulation data to calculate the time-frequency image.

[0060] Specifically, it includes the following sub-steps:

[0061] S2.1. Select sweep angle data at a single frequency from the one-dimensional simulation data and the two-dimensional test data respectively as the initial input. Determine the window size according to the number of sweep angles and perform time-frequency transformation. In this embodiment, a smooth pseudo-Wigner-Ville distribution (SPWVD) is used, which is defined as follows:

[0062]

[0063] In the formula, windows g(u) and h(τ) are both real even functions, and s(t) is the one-dimensional complex data defined in step S1;

[0064] The time-frequency curves obtained by SPWVD have high resolution and focus. During the anechoic chamber scattering field test, the radar turntable rotates at a constant speed, and the scanning angle can correspond to the sampling time series. After time-frequency transformation, the time-frequency characteristics of the scattered electric field correspond to the changes in Doppler frequency over time.

[0065] Step S2.2: After obtaining the time-frequency image, first calculate the overall mean of the image. Set the image edge threshold to 1 / 20 of the overall mean. Compare the mean values ​​on the top and bottom sides from left to right to obtain the positions of the top and bottom edges. Check whether there is oversampling or undersampling based on the edge position. An edge located between one-half and three-quarters of the boundary is normal. An edge below one-half of the boundary is undersampling. An edge above three-quarters of the boundary is oversampling. If there is undersampling or oversampling, according to the sampling theorem, it is necessary to construct suitable input data through interpolation or decimation methods and continue to step S2.1.

[0066] Step S3: Based on the Doppler frequency characteristics, detect the local scattering center, perform IRT transformation on the time-frequency image, and obtain the location and amplitude information of the local scattering center (LSC).

[0067] Step S3 specifically includes the following sub-steps:

[0068] Step S3.1, the expression for the Local Scattering Center (LSC) model is:

[0069]

[0070] In the formula, f is the radar frequency, φ is the visible azimuth angle range of the scattering center, and r i The scattering center position vector, A is the radar's viewpoint direction vector. i The amplitude at the scattering center;

[0071] The Doppler frequency f at the LSC scattering center Di (t) is represented as:

[0072]

[0073] In the formula, ζ represents the radar perspective. With the scattering center position r i The angle between them Let f be the distance of the scattering center from the radar's perspective. From the formula derivation, it can be seen that f... Di It exhibits a sinusoidal variation as ζ changes;

[0074] Step S3.2: Process the time-frequency graph using IRT (Inverse Radon Transform):

[0075]

[0076] In the formula, φ is the radar azimuth angle, and k x =vcosφ,k y =vsinφ, for a single sine curve The phase of the sine curve is represented by the time-frequency graph after IRT. It is represented by a fixed point, with the location being... Extracting the corresponding coordinates (x0, y0), the amplitude and phase of the sine curve corresponding to the LSC can be expressed as:

[0077]

[0078]

[0079] Step S3.3: Obtain all LSC positions of the simulation data and test data respectively. Select and set an appropriate distance difference according to the target to be tested. In this example, it is set to 0.01m. Extract the intersection of LSC positions as the final LSC position. The amplitude is the average of the two, and the visible range is the maximum visible range of the two.

[0080] Step S4: Based on the Doppler frequency characteristics, perform one-dimensional peak detection on each row of the time-frequency image, set the distributed scattering center (DSC) detection threshold and the minimum DSC length, and detect and extract DSC information (visible angle, length, amplitude, and position information).

[0081] Step S4 specifically includes the following sub-steps:

[0082] Step S4.1, the expression for the distributed scattering center (DSC) model is:

[0083]

[0084] In the formula, f is the radar frequency, and φ i L represents the visible angle position of DSC. i r is the DSC length parameter. i The scattering center position vector, A is the radar's viewpoint direction vector. i The amplitude of the scattering center is given. As can be seen from the formula derivation, the DSC appears as a vertical bright line in the time-frequency graph, and only appears at specific viewing angles.

[0085] Step S4.2: On the same column, detect the maximum value information by row. If the extreme value is higher than the set DSC threshold, and the extreme value extracted after the data in the column is close to the extreme value before the column, or the vertical bright line visibility angle of the adjacent column is close, it is considered to be the same feature and merged. Set the minimum DSC length to 2% of the number of data points. Line features with a length lower than this are not considered.

[0086] The difference in Doppler frequencies between the two endpoints of the DSC is:

[0087]

[0088] In the formula, A and B are the two endpoints of the vertical bright line, x and y are the corresponding coordinates, and f D The Doppler frequency is the DSC frequency at the visible angle. and Perpendicular, the line segment can be obtained based on the difference in Doppler frequencies. The actual length L i r Combining the DSC model and the time-frequency transformation formula, the Doppler frequency difference Δf can be obtained. DWith the model's true length L i r and DSC length parameter L i The relationship between them is:

[0089]

[0090] In the formula, L i r L represents the actual length of the model. i For DSC length parameter, For wave vectors;

[0091] Step S4.3: Obtain all DSC angles from the simulation data and the test data respectively, set the angle difference to 0.5°, extract the intersection of the DSC occurrence angles as the final DSC occurrence angle, and use the average of the two for the length.

[0092] Step S5: Based on the appearance angle of the DSC obtained in step S4, combined with the target test data, and based on the DSC length obtained in step S4, obtain the accurate positions of the two endpoints of the DSC through the one-dimensional distance image, and solve for the precise position of the DSC.

[0093] Based on the angle at which the DSC appears in step S4, select the two-dimensional sweep frequency test data at the corresponding angle to form a one-dimensional range image. First, find all the peak points of the one-dimensional range image. According to the order of peak amplitude, compare whether the DSC position conforms to the Doppler frequency range in the time-frequency image. If the distance conforms to the corresponding DSC length, the corresponding one-dimensional range image position is determined to be the two endpoints of the DSC.

[0094] Step S6: Substitute the obtained scattering center information into the scattering center model to obtain the reconstructed scattering field. Then, adjust the model according to the relationship between the time-frequency image amplitude and the electric field amplitude to make the amplitudes of the LSC and DSC models more accurate.

[0095] The RCS is generated based on the LSC scattering center model and compared with the input single-frequency swept RCS. The model accuracy is judged by the variance of the amplitude difference array between the two. Since the time-frequency transformation will affect the electric field amplitude, it is necessary to calculate the ratio of the peak value of the input RCS to the peak value of the ASC reconstructed RCS, and use it as an adjustment coefficient to multiply the amplitude parameter of the ASC model.

[0096] This invention employs a combined computational and measurement method for extracting attribute scattering centers and estimating parameters, which is of great value for the study and analysis of radar target characteristics. Compared with HRRP history maps, SAR imaging, and ISAR imaging, this algorithm reduces the amount of data required for attribute scattering center parameter estimation, improves computational efficiency while ensuring model accuracy, and can effectively improve the accuracy of the reconstructed model's RCS.

[0097] It should be noted that, in the embodiments of the present invention, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the embodiments and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0098] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A method for extracting the scattering center of radar target attributes, characterized in that, Includes the following steps: Step S1: Obtain the one-dimensional single-frequency sweep angle simulation data and two-dimensional sweep frequency sweep angle test data of the target. Based on the far-field radar scattering cross section and phase, the far-field scattering data of the target is uniformly represented as a complex electric field. Step S2: Select one-dimensional single-frequency sweep angle simulation data to calculate the time-frequency image; Step S3: Based on the Doppler frequency characteristics, detect the local scattering center, perform IRT transformation on the time-frequency image, and obtain the location and amplitude information of the local scattering center LSC; Step S4: Based on the Doppler frequency characteristics, one-dimensional peak detection is performed on each row of the time-frequency image. The detection threshold and minimum DSC length of the distributed scattering center DSC are set, and DSC information is detected and extracted. Step S5: Determine the precise position of the DSC based on the appearance angle and length of the DSC obtained in step S4. Step S6: Substitute the obtained scattering center information into the scattering center model to obtain the reconstructed scattering field. Then, adjust the model according to the relationship between the time-frequency image amplitude and the electric field amplitude to make the amplitudes of the local scattering center LSC and the distributed scattering center DSC models more accurate.

2. The radar target attribute scattering center extraction method as described in claim 1, characterized in that, The target far-field scattering data is uniformly represented as a complex electric field. The correspondence between the far-field scattering results and the electric field in the simulation software and testing system is as follows: In the formula, σ The radar cross section (RCS) is expressed in dBsm. Phase, in degrees. j For complex units, E It is a complex electric field.

3. The radar target attribute scattering center extraction method as described in claim 2, characterized in that, Step S2 includes the following steps: S2.

1. Select sweep angle data at a single frequency from the one-dimensional simulation data and the two-dimensional test data respectively as the initial input. Determine the window size according to the number of sweep angles, and use smooth pseudo-Wigner-Ville distribution SPWVD for time-frequency transformation: In the formula, window , All are real even functions. s ( t () represents the one-dimensional complex data defined in step S1; After time-frequency transformation, the time-frequency characteristics of the scattered electric field correspond to the variation characteristics of the Doppler frequency with time. Step S2.2: After obtaining the time-frequency image, first calculate the overall mean of the image. Set the image edge threshold to 1 / 20 of the overall mean. Compare the mean values ​​of the upper and lower sides from left to right to obtain the positions of the upper and lower edges.

4. The radar target attribute scattering center extraction method as described in claim 3, characterized in that, In step S2.2, check whether there is oversampling or undersampling based on the edge position. If the edge is located between one-half and three-quarters of the boundary, it is normal. If the edge is below one-half of the boundary, it is undersampling. If it is above three-quarters, it is oversampling. If there is undersampling or oversampling, the input data needs to be reconstructed by interpolation or extraction methods, and step S2.1 needs to be repeated.

5. The radar target attribute scattering center extraction method as described in claim 3, characterized in that, Step S3 includes the following steps: Step S3.1, the expression for the Local Scattering Center (LSC) model is: In the formula, f For radar frequency, The visible angular range of the azimuth angle of the scattering center. The scattering center position vector, This is the radar's viewpoint direction vector. The amplitude at the scattering center; Doppler frequency at the LSC scattering center Represented as: In the formula, Radar view direction vector With the scattering center position vector The angle between them The distance of the scattering center from the radar's perspective. along with The changes exhibit a sinusoidal pattern. Step S3.2: Process the time-frequency graph using the Radon Inverse Transform (IRT): In the formula, The visible angular range of the azimuth angle of the scattering center. For a single sine curve , The phase of the sine curve is represented by the time-frequency graph after IRT. It is represented by a fixed point, with the location being... Extract the corresponding coordinates Then the amplitude and phase of the sine curve corresponding to LSC are expressed as: Step S3.3: Obtain all LSC positions of the simulation data and test data respectively. Select the distance difference according to the target to be tested, extract the intersection of LSC positions as the final LSC position, use the average of the two values ​​for amplitude, and use the maximum visible range of the two values ​​for visible range.

6. The radar target attribute scattering center extraction method as described in claim 5, characterized in that, Step S4 includes the following steps: Step S4.1, the model expression for the distributed scattering center DSC is: In the formula, f For radar frequency, Indicates the visible angle position of DSC. For DSC length parameter, The scattering center position vector, This is the radar's viewpoint direction vector. The amplitude of the scattering center is represented by a vertical bright line in the time-frequency graph; Step S4.2: On the same column, detect the maximum value information by row. If the extreme value is higher than the set DSC threshold, and the extreme value extracted after the data in the column is close to the extreme value before the column, or the vertical bright line visibility angle of the adjacent column is close, it is considered to be the same feature and merged. Set the minimum DSC length to 2% of the number of data points. Line features with a length lower than this are not considered. The difference in Doppler frequencies between the two endpoints of the DSC is: In the formula, a and b are the two endpoints of the vertical bright line, and x and y are the corresponding coordinates. The Doppler frequency is the DSC frequency at the visible angle. and Perpendicular, the line segment is obtained based on the difference in Doppler frequencies. The actual length Combining the DSC model and the time-frequency transformation formula, the difference in Doppler frequencies is... With the model's true length and DSC length parameter The relationship between them is: In the formula, For the actual length of the model, For DSC length parameter, For wave vectors; Step S4.3: Obtain all DSC angles from the simulation data and the test data respectively, set the angle difference to 0.5°, extract the intersection of the DSC occurrence angles as the final DSC occurrence angle, and use the average of the two for the length.

7. The radar target attribute scattering center extraction method as described in claim 6, characterized in that, In step S5, based on the angle at which the DSC appears in step S4, two-dimensional sweep frequency test data at the corresponding angle are selected to form a one-dimensional distance image. All peak points of the one-dimensional distance image are found. According to the order of peak amplitude, the DSC position is compared to see if it matches the Doppler frequency range in the time-frequency image. If the distance matches the corresponding DSC length, the corresponding one-dimensional distance image position is determined to be the two endpoints of the DSC.

8. The radar target attribute scattering center extraction method as described in claim 7, characterized in that, In step S6, the RCS is generated based on the scattering center model LSC and compared with the input single-frequency sweep RCS. The model accuracy is judged by the variance of the amplitude difference array between the two. Since the time-frequency transformation will affect the electric field amplitude, it is necessary to calculate the ratio of the peak value of the input RCS to the peak value of the ASC reconstructed RCS, and use it as an adjustment coefficient to multiply the amplitude parameter of the ASC model.

Images