Shipborne ground wave radar array pattern error calibration method based on multi-AIS information and target time-frequency characteristics

Abstract

The invention relates to a shipborne ground wave radar array directional diagram error calibration method based on multi-AIS information and target time-frequency characteristics. The method comprises the following steps: screening out a plurality of reference signal sources meeting requirements from radar time domain data, AIS information and attitude information of a shipborne platform in a time-frequency dimension, wherein the radar time domain data, the AIS information and the attitude information are obtained in heading change time; extracting time-frequency information of each reference signal source to estimate an error matrix in a single azimuth range; combining a plurality of azimuth ranges to construct an error matrix in a radar detection range; and compensating an array output signal to calibrate an array directional diagram. According to the method, AIS information of a plurality of ship targets is utilized, target azimuth angle changes caused by ship-borne platform yawing characteristics and time-frequency analysis processing of the targets are combined, time-frequency data of the targets are extracted to estimate directional diagram errors, calibration of the directional diagram errors is achieved, and the direction finding performance of the targets is improved.

Description

Technical Field

[0001] This invention relates to a method for calibrating the pattern error of a shipborne ground wave radar, specifically a method for calibrating the pattern error of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics. Background Technology

[0002] High-Frequency Surface Wave Radar (HFSWR) utilizes the diffraction of high-frequency (3~30MHz) vertically polarized electromagnetic waves along the coastal plane to detect and track targets such as ships and low-altitude aircraft at sea. It offers advantages such as wide coverage, beyond-line-of-sight capability, and continuous detection. Compared to shore-based ground wave radar, shipborne HFSWR has its transmitting and receiving stations mounted on mobile platforms, leveraging the platform's mobility and flexibility for a wider detection range. However, when estimating the Direction of Arrival (DOA) of a target, shipborne HFSWR is affected by the surrounding environment and platform motion, leading to distortion in the array radiation pattern and a decrease in DOA estimation accuracy. Therefore, to improve the direction-finding performance of shipborne HFSWR, it is necessary to consider the impact of the surrounding environment and platform motion on the radiation pattern, develop a radiation pattern error calibration method suitable for shipborne systems, correct errors, ensure data accuracy, and thus improve direction-finding precision.

[0003] Current research on pattern error calibration for shipborne ground-wave radar is relatively limited, with most studies focusing on shore-based ground-wave radar. One existing method utilizes Automatic Identification System (AIS) information as an auxiliary source for pattern error calibration. This involves accumulating AIS information from various azimuths and ship echo signals to estimate and calibrate the error. However, this method requires accumulating AIS information from all azimuths of interest to obtain the true azimuth of the ship echoes, resulting in a long implementation cycle and high cost. In reality, shipborne ground-wave radar often has limited AIS information in some detection areas, making it difficult to acquire a large number of high signal-to-noise ratio ship echoes. Furthermore, due to the motion characteristics of shipborne radar, the surrounding environment is constantly changing, causing the error value to vary over time, thus requiring real-time calibration. Therefore, traditional shore-based calibration methods are not suitable for shipborne systems. This invention utilizes AIS information from multiple ship targets, combined with the azimuth changes of the targets in the radar coordinate system caused by the yaw characteristics of the shipborne platform, and the time-frequency analysis and processing of the targets, to extract the broadened time-frequency data of the targets to estimate the radiation pattern error, thereby achieving the calibration of the radiation pattern error and improving the direction finding performance of the targets. Summary of the Invention

[0004] (a) The technical problems to be solved.

[0005] The present invention aims to provide a method for calibrating the radiation pattern of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics, so as to improve the direction finding accuracy of ship targets.

[0006] (ii) Technical solution.

[0007] This invention includes the following steps:

[0008] Step 1: Acquisition of data within the heading change.

[0009] The radar time-domain data, AIS information, and shipboard platform attitude information are acquired during the heading change time. The time-domain data is transformed to the time-frequency domain to obtain the radar time-frequency (TF). Then, the target latitude, longitude, speed, and other information provided by the platform's attitude information and AIS information are projected onto the time-frequency and matched with the target information.

[0010] Step 2: Screening of multiple reference signal sources and extraction of time and frequency information.

[0011] For the matched target, the precise azimuth information of each sampling point within the accumulation time is calculated based on the platform's attitude information and the latitude and longitude provided by AIS information. ;according to N targets meeting the following requirements are selected: the combined azimuth range of each target should cover as much of the radar detection area as possible (-60°~60°). Then, the time-frequency information of each target is extracted: based on the energy intensity of each time unit, the entire time unit is divided into several levels, and different frequency cells are extracted for each level. This method is used to extract the target time-frequency information for each channel.

[0012] Step 3: Estimation of the error matrix within a single azimuth range.

[0013] Based on the target's time-frequency information obtained in step 2, a covariance matrix is ​​constructed for the time-frequency information at each time point, and eigenvalue decomposition is performed to obtain the signal subspace; simultaneously, the azimuth information obtained in step 2 is used... Construct an ideal guidance vector; combine the signal subspace and orientation information. Calculate the error value at each time point (i.e., different orientations) to construct an error matrix. ;

[0014] Step 4: Construct the error matrix across multiple azimuth ranges.

[0015] Step 3 processes the time-frequency information of each target obtained in step 2 to obtain the information within each azimuth range. This constitutes the overall error matrix. .

[0016] To address the potential overlap between two azimuth ranges, the average error value within the overlapping range is used to represent the overlapping portion. This process is repeated for all overlapping portions to obtain the updated error matrix. .

[0017] Step 5: Array data compensation.

[0018] Using the error matrix obtained in step 4, the array output data is compensated to obtain the calibrated radiation pattern.

[0019] (iii) Beneficial effects.

[0020] The advantages of this invention are as follows:

[0021] This invention utilizes AIS information from multiple ship targets, combined with the azimuth changes of the targets in the radar coordinate system caused by the yaw characteristics of the shipborne platform, and the time-frequency analysis and processing of the targets, to extract the broadened time-frequency data of the targets to estimate the radiation pattern error, thereby achieving the calibration of the radiation pattern error and improving the direction finding performance of the targets. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the basic process of the present invention.

[0023] Figure 2 This refers to the range of azimuth angle changes for multiple targets under varying heading conditions according to the present invention.

[0024] Figure 3 The time-frequency spectrum diagrams of target channels 1 to 4 of this invention are shown.

[0025] Figure 4 The time-frequency spectrum of the target channels 5-8 of this invention is shown.

[0026] Figure 5 This is a comparison of the radiation patterns before and after calibration in this invention. Detailed Implementation

[0027] To make the objectives, contents, and advantages of the present invention clearer, the following description is provided in conjunction with the appendix. Figure 1 The specific embodiments of the present invention will be described in further detail below.

[0028] Step 1: Acquisition of data within the heading change.

[0029] Changes in azimuth of multiple targets under changing heading conditions, such as Figure 2 As shown, the bow direction of the shipborne platform has changed. Then the azimuth angles of multiple targets within the detection range also change in the radar coordinate system. There may be cases where the ranges of change for two targets overlap. The radar time-domain data for the bow change time is obtained. The data includes AIS information and the attitude information of the shipborne platform, where M is the number of array elements. The time-domain data is transformed to the time-frequency domain using the Short-time Fourier Transform (STFT) method to obtain the radar time-frequency (TF). It is assumed that the target signal of the shipborne ground wave radar is a single-component signal.

[0030] ,

[0031] In the formula, It is time. For instantaneous amplitude, This is the instantaneous phase. Therefore... The STFT time spectrum is:

[0032] ,

[0033] In the formula, For frequency, To analyze the signal, To move the window.

[0034] Figure 3 The time-frequency spectrum of a certain target in channels 1-4. Figure 4 The time-frequency spectra are for channels 5 to 8.

[0035] Simultaneously, the platform's attitude information and the target's latitude, longitude, speed, and other information provided by AIS are projected onto the time spectrum and matched with the target information:

[0036] ,

[0037] in This represents the target's velocity calculated from AIS information and platform attitude information. This represents the velocity of a target in the radar time spectrum, when the absolute value of the difference between the two is less than a certain threshold. When the time is right, it means the match is successful. It is related to the current radar velocity resolution.

[0038] Step 2: Screening of multiple reference signal sources and extraction of time and frequency information.

[0039] For the matched target, the precise azimuth information of each sampling point within the accumulation time is calculated based on the platform's attitude information and the latitude and longitude provided by AIS information. Where T is the accumulation time; according to Select N targets that meet the following requirements: Each target The combined range should cover as much of the radar detection area as possible (-60°~60°). Then, the time-frequency information of each target is extracted: based on the energy intensity of each time unit, the entire time unit is divided into several levels, and different frequency cells are extracted from each level. This method is used to extract the target time-frequency information for each channel. Let the energy of the entire time dimension be... Calculate the 25th, 50th, and 75th percentiles. The entire time dimension is divided into four levels based on quantiles, and each level of the time dimension is labeled according to its energy intensity:

[0040] ,

[0041] Extracting based on the frequency dimension By widening the grid cells, the target widening information was fully utilized, and the time-frequency data of each corresponding channel was extracted. .

[0042] Step 3: Estimation of the error matrix within a single azimuth range.

[0043] Based on the time-frequency information of the target obtained in step 2, a covariance matrix is ​​constructed for the time-frequency information at each time point, and eigenvalue decomposition is performed:

[0044] ,

[0045] ,

[0046] In the formula, The eigenvector corresponding to the largest eigenvalue after constructing the covariance matrix and eigenvalue decomposition of the target time-frequency data represents the signal subspace of the signal.

[0047] Simultaneously utilize the orientation information obtained in step 2. Construct the ideal guide vector in each direction:

[0048] ,

[0049] in,

[0050] ,

[0051] In the above formula, For carrier frequency, , The distance between array elements.

[0052] joint and location information Calculate the error value at each time point (i.e., different orientations). This constitutes the error matrix. They satisfy the following relationship:

[0053] ,

[0054] in, Therefore, the error value can be expressed as:

[0055] ,

[0056] The error matrix is: .

[0057] Step 4: Construct the error matrix across multiple azimuth ranges.

[0058] Step 3 processes the time-frequency information of each target obtained in step 2 to obtain the information within each azimuth range. This constitutes the overall error matrix. .

[0059] Regarding the potential overlap between two directional ranges, i.e., the two targets... The ranges may overlap, resulting in the error matrix. There will also be overlapping parts, for two The overlapping part is defined as and ,right and Taking the average yields:

[0060] ,

[0061] use To replace two The overlapping portions are then processed as described above. This process is applied to all overlapping portions to obtain the updated error matrix. .

[0062] Step 5: Array data compensation.

[0063] Using the error matrix obtained in step 4, the array output data is compensated to obtain the calibrated radiation pattern, such as... Figure 5 As shown, the diagram illustrates the comparison of radiation patterns before and after calibration within a certain azimuth range. It can be seen from the diagram that after adding the estimated error matrix to the ideal radiation pattern, it closely matches the uncalibrated radiation pattern. Similarly, after compensating the array data with the estimated error, it closely matches the ideal radiation pattern.

Claims

1. A method for calibrating the pattern error of a shipborne ground-wave radar array based on multi-AIS information and target time-frequency characteristics, characterized in that, include: (1) Acquisition of data during heading change: acquire radar time-domain data, AIS information and attitude information of the shipborne platform during heading change; transform the time-domain data to the time-frequency domain to obtain the radar time-frequency (TF); then use the platform's attitude information and the target latitude, longitude, speed and other information provided by AIS information to project it into the time-frequency and match it with the target information. (2) Screening of multiple reference signal sources and extraction of time and frequency information: For the matched target, the precise azimuth information of each sampling point within the accumulation time is calculated based on the attitude information of the platform and the latitude and longitude provided by the AIS information. ;according to N targets that meet the following requirements are selected: the joint azimuth range of each target should cover the azimuth range of interest in the radar detection area (-60°~60°); then the time and frequency information of each target is extracted: according to the energy intensity of each time unit, the entire time unit is divided into several levels, and different frequency cells are extracted for each level. This method is used to extract the target time and frequency information for each channel. (3) Error matrix estimation within a single azimuth range: Based on the obtained time-frequency information of the target, a covariance matrix is ​​constructed for the time-frequency information at each time point, and eigenvalue decomposition is performed to obtain the signal subspace; simultaneously, azimuth information is utilized. Construct an ideal guidance vector; combine the signal subspace and orientation information. Calculate the error values ​​for different orientations at each time point to construct an error matrix. ; (4) Construction of error matrices within multiple azimuth ranges: For each target's time-frequency information, error matrix estimation is performed within a single azimuth range to obtain the error matrix within each azimuth range. This constitutes the overall error matrix. To address the potential overlap between two directional ranges, the average error value within the overlapping range is used to represent the overlapping portion. This process is repeated for all overlapping portions to obtain the updated error matrix. ; (5) Array data compensation: The obtained error matrix is ​​used to compensate the array output data to obtain the calibrated radiation pattern.

2. The method for calibrating the pattern of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics as described in claim 1, characterized in that, Acquisition of data within the heading change includes: If the bow direction of the ship's platform changes Then the azimuth angles of multiple targets within the detection range also change in the radar coordinate system. There may be cases where the range of change of two targets overlaps; obtain radar time-domain data within the time of heading change. The data includes AIS information and the attitude information of the shipborne platform, where M is the number of array elements. The time-domain data is transformed to the time-frequency domain to obtain the radar time-frequency (TF). It is assumed that the target signal of the shipborne ground wave radar is a single-component signal. ， In the formula, It is time. For instantaneous amplitude, If it is the instantaneous phase, then The STFT time spectrum is: ， In the formula, For frequency, To analyze the signal, To move the window; Simultaneously, the platform's attitude information and the target's latitude, longitude, speed, and other information provided by AIS are projected onto the time spectrum and matched with the target information: ， in This represents the target's velocity calculated from AIS information and platform attitude information. This represents the velocity of a target in the radar time spectrum, when the absolute value of the difference between the two is less than a certain threshold. When the time is right, it means the match is successful. It is related to the current radar velocity resolution.

3. The method for calibrating the pattern of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics according to claim 1, characterized in that, Screening of multiple reference signal sources and extraction of time-frequency information, including: For the matched target, the precise azimuth information of each sampling point within the accumulation time is calculated based on the platform's attitude information and the latitude and longitude provided by AIS information. Where T is the accumulation time; according to Select N targets that meet the following requirements: Each target The combined range should cover the azimuth range of interest within the radar detection area (-60°~60°); then, the time-frequency information of each target is extracted: based on the energy intensity of each time unit, the entire time unit is divided into several levels, and different frequency cells are extracted from each level. This method is used to extract the target time-frequency information for each channel. Let the energy of the entire time dimension be... Calculate the 25th, 50th, and 75th percentiles. The entire time dimension is divided into four levels based on quantiles, and the level of each time dimension is labeled according to the strength of energy: ， Extracting based on the frequency dimension By widening the grid cells, the target widening information was fully utilized, and the time-frequency data of each corresponding channel was extracted. .

4. The method for calibrating the pattern error of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics as described in claim 1, characterized in that, Error matrix estimation within a single orientation range includes: Based on the obtained time-frequency information of the target, a covariance matrix is ​​constructed for the time-frequency information at each time point, and eigenvalue decomposition is performed: ， ， In the formula, The eigenvector corresponding to the largest eigenvalue after constructing the covariance matrix and eigenvalue decomposition of the target time-frequency data represents the signal subspace of the signal. Simultaneously utilize location information Construct the ideal guide vector in each direction: ， in, ， In the above formula, For carrier frequency, , The spacing between array elements; joint and location information Calculate the error value for different orientations at each time point. This constitutes the error matrix. They satisfy the following relationship: ， in, Therefore, the error value can be expressed as: ， The error matrix is: .

5. The method for calibrating the pattern error of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics according to claim 1, characterized in that, The construction of error matrices across multiple azimuth ranges includes: For each target's obtained time-frequency information, an error matrix estimation is performed within a single azimuth range to obtain the error matrix within each azimuth range. This constitutes the overall error matrix. ; Regarding the potential overlap between two directional ranges, i.e., the two targets... The ranges may overlap, resulting in the error matrix. There will also be overlapping parts, for two The overlapping part is defined as and ,right and Taking the average yields: ， use To replace two The overlapping portions; by performing the above processing on all overlapping portions, the updated error matrix can be obtained. .

6. The method for calibrating the pattern error of a shipborne ground wave radar array based on multiple AIS information and target time-frequency characteristics according to claim 1, characterized in that, Array data compensation, including: By using the obtained error matrix to compensate for the array output data, a calibrated radiation pattern can be obtained.

Images