Radar signal automatic analysis warehousing method and device based on characteristic vector

Abstract

The invention discloses a radar signal automatic analysis warehousing method and device based on a characteristic vector, and belongs to the field of radar signal processing. The method comprises the following steps: reading a target radar signal, and extracting a plurality of valid data segments by using a preset invalid data time window as a segmentation standard; sequentially inputting the effective data segments into a characteristic vector generation model to obtain a characteristic vector corresponding to each effective data segment; calculating the correlation of each effective data segment by using a correlation algorithm, performing grouping clustering on each effective data segment based on the correlation, and forming a correlation feature vector matrix by using the feature vectors of the effective data segments in each group; and on the basis of the correlation feature vector matrix, correspondingly storing each valid data segment into a relational database. According to the scheme, autonomous learning and fusion grouping can be carried out on the radar signals, the correlation feature vector matrix is established and then stored in the relational database, and compared with manual identification and screening, the sorting efficiency and accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar signal processing, and in particular to a method and device for automatically analyzing and storing radar signals based on characteristic vectors. Background Art

[0002] In existing radar signal processing, it is already possible to obtain excellent one-dimensional images (lines), two-dimensional images (graphs) based on the data source, and even use time-dependent two-dimensional images to generate three-dimensional images (volumes). Data are generally stored in files, and each file contains a lot of unstructured information about signal characteristics, which is very inconvenient to use.

[0003] However, unlike general signal processing, the characteristic information in the radar field contains many unknown or unidentified signals. Manual identification and screening of these signals is time-consuming and labor-intensive, and there is also a risk of errors and omissions.

[0004] Therefore, there is an urgent need to provide a method and device for automatically analyzing and storing radar signals based on characteristic vectors. Summary of the Invention

[0005] In order to solve the problem that radar signals require manual identification and screening, which is time-consuming and labor-intensive and prone to errors and omissions, an embodiment of the present invention provides a method and device for automatic analysis and storage of radar signals based on characteristic vectors.

[0006] On the one hand, a method for automatically analyzing and storing radar signals based on characteristic vectors is provided, the method comprising:

[0007] Read the target radar signal and use the pre-set invalid data time window as the segmentation standard to extract several valid data segments;

[0008] Inputting the valid data segments into a feature vector generation model in sequence to obtain a feature vector corresponding to each valid data segment;

[0009] Calculating the correlation of each of the valid data segments using a correlation algorithm, so as to group and cluster the valid data segments based on the correlation, and constructing a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group;

[0010] Based on the correlation feature vector matrix, each valid data segment is correspondingly stored in a relational database.

[0011] On the other hand, a device for automatically analyzing and storing radar signals based on characteristic vectors is provided, which is used to implement the steps described in any method embodiment of the specification, and the device includes:

[0012] The extraction unit is used to read the target radar signal and extract several valid data segments using the pre-set invalid data time window as a segmentation criterion;

[0013] A generating unit, configured to sequentially input the valid data segments into a feature vector generating model to obtain a feature vector corresponding to each valid data segment;

[0014] a grouping unit, configured to calculate the correlation of each of the valid data segments using a correlation algorithm, so as to group and cluster the valid data segments based on the correlation, and to construct a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group;

[0015] The storage unit is used to store each valid data segment in a relational database based on the correlation feature vector matrix.

[0016] On the other hand, a computer device is provided, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of the above-mentioned method.

[0017] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored in the storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method are implemented.

[0018] On the other hand, a computer program product is provided, comprising a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0019] The technical solution provided by the present invention can at least bring the following beneficial effects:

[0020] First, through the invalid data time window, several valid data segments are extracted from the target radar signal, and invalid data is eliminated. Then, the feature vector generation model is used to learn and generate the feature vectors of the valid data segments. Then, the valid data segments are grouped and clustered using a correlation algorithm, and a correlation feature vector matrix is established and stored in a relational database. Therefore, this solution can autonomously learn and fuse radar signals, group them, and establish a correlation feature vector matrix before storing it in a relational database. It can automatically organize data feature information for researchers, converting unstructured data files into feature vector information and storing it in a relational database. Compared with manual identification and screening, it can improve the efficiency and accuracy of the organization, and provide a basic data set for subsequent large-scale model learning and other scientific research fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 This is a flow chart of a method for automatically analyzing and storing radar signals based on characteristic vectors provided by one embodiment of the present invention;

[0023] Figure 2 This is a structural diagram of a radar signal automatic analysis and warehousing device based on characteristic vectors provided by one embodiment of the present invention;

[0024] Figure 3 This is a hardware architecture diagram of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] The specific implementation of the above concept is described below.

[0027] Please refer to Figure 1 An embodiment of the present invention provides a method for automatically analyzing and storing radar signals based on characteristic vectors, the method comprising:

[0028] Step 100: Read the target radar signal and use the preset invalid data time window as a segmentation criterion to extract several valid data segments;

[0029] Step 102: Input the valid data segments into the feature vector generation model in sequence to obtain the feature vector corresponding to each valid data segment;

[0030] Step 104: Calculate the correlation of each valid data segment using a correlation algorithm, group and cluster the valid data segments based on the correlation, and construct a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group;

[0031] Step 106: Based on the correlation feature vector matrix, store each valid data segment in a relational database.

[0032] In an embodiment of the present invention, several valid data segments are first extracted from the target radar signal using an invalid data time window, and invalid data is discarded. A feature vector generation model is then used to learn and generate feature vectors for the valid data segments. A correlation algorithm is then used to group and cluster the valid data segments, and a correlation feature vector matrix is established and stored in a relational database. Therefore, this solution can autonomously learn and fuse radar signals, group them, and establish a correlation feature vector matrix for storage in a relational database. It can automatically organize data feature information for researchers, converting unstructured data files into feature vector information and storing it in a relational database. Compared to manual identification and screening, this approach can improve organization efficiency and accuracy, and can provide a basic data set for subsequent large-scale model learning and other scientific research fields.

[0033] Described below Figure 1 How to perform the steps shown.

[0034] For step 100:

[0035] In some embodiments, before reading the target radar signal, the method further includes:

[0036] The original radar signal is preprocessed using fast Fourier transform to obtain the target radar signal.

[0037] In this embodiment, during the radar data acquisition process, fast Fourier transform is used to pre-process the data to obtain a data stream after noise removal.

[0038] In some embodiments, step 100 reads the target radar signal and uses a preset invalid data time window as a segmentation criterion to extract a number of valid data segments, which may include:

[0039] Obtaining an interval threshold of invalid data, where the interval threshold is used to determine data exceeding the interval threshold as valid data;

[0040] Read target radar signals and smoothly scroll through invalid data time windows;

[0041] When valid data is encountered, continuous extraction is performed until the amount of invalid data read continuously is greater than the invalid data time window, and a valid data segment is obtained, thereby extracting several valid data segments from the target radar signal.

[0042] In this embodiment, the interval threshold of invalid data is a positive and negative interval threshold near zero. Data exceeding or below the threshold is valid data. Data between the thresholds belongs to the noise floor signal and is invalid data.

[0043] This embodiment sets a time window for invalid signals in the time dimension. Each valid data segment must be preceded and followed by a noise floor signal with a length greater than the time window. Therefore, during the extraction of each valid data segment, the first data point above or below the threshold is considered a valid signal, and valid data is read continuously until the amount of invalid data read exceeds the time window, resulting in a valid data segment.

[0044] In some implementations, after the amount of invalid data continuously read exceeds the invalid data time window and before obtaining a valid data segment, the method further includes:

[0045] The invalid data in the invalid data time window is eliminated to extract the radar signal between the first valid data extracted this time and the previous valid data in the invalid data time window to obtain a valid data segment.

[0046] In this embodiment, in order to accurately extract the valid signal of the valid data segment, after the amount of invalid data read continuously is greater than the invalid data time window, the invalid data in the invalid data time window can be eliminated, so as to extract the radar signal between the first valid data extracted this time and the previous data of the invalid data time window, and obtain an accurate and effective valid data segment.

[0047] Regarding step 102:

[0048] In some implementations, step 102 sequentially inputs the valid data segments into a feature vector generation model to obtain a feature vector corresponding to each valid data segment, including:

[0049] For each valid data segment, execute:

[0050] Calculate the MD5 value of the current valid data segment and collect the label information of the current valid data segment;

[0051] Save the current valid data segment into a file, and use the MD5 value and tag information of the current valid data segment as the file name;

[0052] The file corresponding to the current valid data segment and its file name are input into the feature vector generation model to obtain the feature vector of the current valid data segment.

[0053] In this embodiment, in existing radar signal processing methods, radar data is primarily stored in the form of files. Data characteristics, such as band and polarization, are not stored with the data files, making it difficult to accurately identify valid information within the data. The absence of these characteristic parameters also increases the complexity of radar data retrieval and querying. Therefore, traditional radar signal storage methods are particularly crude, relying entirely on manual identification and screening. In this embodiment, to improve the classifiability and effectiveness of radar signals, the MD5 value of the current valid data segment is calculated, and tag information for the current valid data segment, including polarization direction, band, frequency, and other information, is collected to generate a file name for the current valid data segment. After obtaining the file and file name corresponding to the valid data segment, the file is input into a feature vector generation model to obtain a feature vector for the current valid data segment. Compared to directly generating a feature vector from the valid data segment, the feature vector obtained in this embodiment is more accurate and effective.

[0054] Regarding steps 104 and 106:

[0055] In some embodiments, step 104 uses a correlation algorithm to calculate the correlation of each valid data segment to group and cluster each valid data segment based on the correlation, including:

[0056] For each valid data segment, execute:

[0057] Calculate the correlation between the current valid data segment and the valid data segments in each group after grouping using a correlation algorithm;

[0058] If the correlation between the current valid data segment and each valid data segment in one of the groups is greater than a set threshold, the current valid data segment is assigned to the group;

[0059] If the current valid data segment does not meet the correlation requirement with each group, the current valid data segment becomes a new group.

[0060] In this embodiment, a correlation algorithm is used to calculate the correlation between the current valid data segment and the valid data segments in each grouped group, so as to group and cluster the data segments based on the correlation and form several correlation feature vector matrices with data of similar characteristics.

[0061] By repeating the above steps, you can automatically establish a eigenvector matrix separated by noise floor time intervals for any radar signal, and store the files containing the corresponding valid data segments and their file names in the relational database in sequence based on the correlation eigenvector matrix, thus converting the unstructured files into structured files, which include not only the original valid signals, but also eigenvectors, MD5 values, and polarization directions, bands, frequencies and other information, providing a high-quality data foundation for subsequent data processing using AI, large models and other methods.

[0062] Please refer to Figure 2The embodiment of the present invention provides a device for automatically analyzing and storing radar signals based on characteristic vectors, which is used to implement the steps of any method embodiment in the specification. The device includes:

[0063] The extraction unit 201 is used to read the target radar signal and extract a number of valid data segments using a preset invalid data time window as a segmentation criterion;

[0064] A generating unit 202 is configured to sequentially input valid data segments into a feature vector generating model to obtain a feature vector corresponding to each valid data segment;

[0065] The grouping unit 203 is configured to calculate the correlation of each valid data segment using a correlation algorithm, so as to group and cluster each valid data segment based on the correlation, and to construct a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group;

[0066] The storage unit 204 is used to store each valid data segment in a relational database based on the correlation feature vector matrix.

[0067] In one embodiment of the present invention, the extraction unit 201 is configured to perform:

[0068] Obtaining an interval threshold of invalid data, where the interval threshold is used to determine data exceeding the interval threshold as valid data;

[0069] Read target radar signals and smoothly scroll through invalid data time windows;

[0070] When valid data is encountered, continuous extraction is performed until the amount of invalid data read continuously is greater than the invalid data time window, and a valid data segment is obtained, thereby extracting several valid data segments from the target radar signal.

[0071] In one embodiment of the present invention, after the amount of invalid data continuously read exceeds the invalid data time window and before obtaining a valid data segment, the extraction unit 201 is further used to: eliminate the invalid data in the invalid data time window, so as to extract the radar signal between the first valid data extracted this time and the previous valid data in the invalid data time window, and obtain a valid data segment.

[0072] In one embodiment of the present invention, the generating unit 202 is configured to execute:

[0073] For each valid data segment, execute:

[0074] Calculate the MD5 value of the current valid data segment and collect the label information of the current valid data segment;

[0075] Save the current valid data segment into a file, and use the MD5 value and tag information of the current valid data segment as the file name;

[0076] The file corresponding to the current valid data segment and its file name are input into the feature vector generation model to obtain the feature vector of the current valid data segment.

[0077] In one embodiment of the present invention, when the grouping unit 203 calculates the correlation of each valid data segment using a correlation algorithm to group and cluster each valid data segment based on the correlation, it is configured to:

[0078] For each valid data segment, execute:

[0079] Calculate the correlation between the current valid data segment and the valid data segments in each group after grouping using a correlation algorithm;

[0080] If the correlation between the current valid data segment and each valid data segment in one of the groups is greater than a set threshold, the current valid data segment is assigned to the group;

[0081] If the current valid data segment does not meet the correlation requirement with each group, the current valid data segment becomes a new group.

[0082] It should be noted that the aforementioned embodiment of the automatic analysis and storage device for radar signals based on characteristic vectors is merely illustrative of the division of the aforementioned functional units. In actual applications, the aforementioned functions can be assigned to different functional units as needed, i.e., the internal structure of the device can be divided into different functional units to perform all or part of the functions described above. Furthermore, the aforementioned device embodiment and the method embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0083] The embodiment of the present application also provides a computer device, please refer to Figure 3 The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement the characteristic vector-based radar signal automatic analysis and storage method provided in the above-mentioned method embodiments.

[0084] An embodiment of the present application also provides a computer-readable storage medium, on which is stored at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the characteristic vector-based radar signal automatic analysis and storage method provided in the above-mentioned method embodiments.

[0085] An embodiment of the present application also provides a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes any of the characteristic vector-based radar signal automatic analysis and warehousing methods in the above embodiments.

[0086] For the convenience of description, the above systems or devices are described as being divided into various modules or units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0087] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that the present application can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application or certain parts of the embodiments.

[0088] Finally, it should be noted that, in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0089] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A radar signal automatic analysis and storage method based on characteristic vectors, characterized in that: include: Read the target radar signal and use the pre-set invalid data time window as the segmentation standard to extract several valid data segments; Inputting the valid data segments into a feature vector generation model in sequence to obtain a feature vector corresponding to each valid data segment; Calculating the correlation of each of the valid data segments using a correlation algorithm, so as to group and cluster the valid data segments based on the correlation, and constructing a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group; Based on the correlation feature vector matrix, each valid data segment is correspondingly stored in a relational database.

2. The method according to claim 1, wherein The target radar signal is read, and a pre-set invalid data time window is used as a segmentation criterion to extract a number of valid data segments, including: Obtaining an interval threshold of invalid data, wherein the interval threshold is used to determine data exceeding the interval threshold as valid data; Reading the target radar signal and smoothly scrolling the invalid data time window; When valid data is encountered, continuous extraction is performed until the amount of invalid data continuously read is greater than the invalid data time window, thereby obtaining a valid data segment, thereby extracting several valid data segments from the target radar signal.

3. The method according to claim 2, wherein After the amount of invalid data continuously read exceeds the invalid data time window and before a valid data segment is obtained, the method further includes: The invalid data in the invalid data time window is removed, and the radar signal between the first valid data extracted this time and the previous valid data in the invalid data time window is extracted to obtain a valid data segment.

4. The method according to claim 1, wherein The step of sequentially inputting the valid data segments into a feature vector generation model to obtain a feature vector corresponding to each valid data segment includes: For each valid data segment, execute: Calculate the MD5 value of the current valid data segment and collect the label information of the current valid data segment; The current valid data segment is stored in a file, and the MD5 value of the current valid data segment and the tag information are used as the file name; The file corresponding to the current valid data segment and its file name are input into the feature vector generation model to obtain the feature vector of the current valid data segment.

5. The method according to claim 1, wherein The calculating the correlation of each valid data segment by using a correlation algorithm to group and cluster each valid data segment based on the correlation includes: For each valid data segment, execute: Calculate the correlation between the current valid data segment and the valid data segments in each group after grouping using a correlation algorithm; If the correlation between the current valid data segment and each valid data segment in one of the groups is greater than a set threshold, the current valid data segment is assigned to the group; If the current valid data segment does not meet the correlation requirement with each group, the current valid data segment becomes a new group.

6. A radar signal automatic analysis and storage device based on characteristic vectors, used to implement the steps of any one of the methods described in claims 1 to 5, characterized in that: include: The extraction unit is used to read the target radar signal and extract several valid data segments using the pre-set invalid data time window as a segmentation criterion; A generating unit, configured to sequentially input the valid data segments into a feature vector generating model to obtain a feature vector corresponding to each valid data segment; a grouping unit, configured to calculate the correlation of each of the valid data segments using a correlation algorithm, so as to group and cluster the valid data segments based on the correlation, and to construct a correlation feature vector matrix using the characteristic vectors of the valid data segments in each group; The storage unit is used to store each valid data segment in a relational database based on the correlation feature vector matrix.

7. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of any one of the methods described in claims 1-5.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that The method comprises a computer program, which implements the steps of the method according to any one of claims 1 to 5 when the computer program is executed by a processor.

Images