Time series-based electronic target collaborative analysis and fusion method and device

Abstract

The application discloses a time series-based electronic target cooperative analysis fusion method and device, which comprises the following steps: acquiring electromagnetic data of an electronic target, performing standardization conversion on electromagnetic characteristics of the electromagnetic data to obtain standardized electromagnetic data, clustering the standardized electromagnetic data to obtain a clustering result, quantifying the clustering result to obtain discretization state data of the electronic target, dividing the discretization state data of the electronic target into multiple sub-items, calculating correlation coefficients between the sub-items based on the multiple sub-items to obtain a cooperative set of the electronic target, and comparing cooperative sets of different electronic targets to determine cooperative relationships between the electronic targets. The method can convert multi-feature electromagnetic data into single-feature discretization electromagnetic state, better quantize data characteristics, and greatly reduce the time complexity of cooperative calculation.

Description

Technical Field

[0001] This application relates to the field of electromagnetic technology, and in particular to a time-series-based method and apparatus for collaborative analysis and fusion of electronic targets. Background Technology

[0002] Electronic targets are defined as signal source devices capable of emitting electromagnetic signals. Electromagnetic detection systems collect electromagnetic data from electronic targets at fixed time intervals, and the collected data conforms to the category of time series. Collaborative analysis analyzes and judges the collaborative working patterns among electronic targets, which helps in intelligence mining and knowledge management of electronic targets. Electronic target time series collaborative analysis is an application scenario of time series analysis. It uses algorithms to analyze the electromagnetic data of electronic targets to determine whether there are collaborative patterns among them.

[0003] Currently, commonly used algorithms for determining collaborative relationships include the Apriori algorithm (hereinafter referred to as the AP algorithm), which mines association rules by generating frequent itemsets. Frequent itemsets are quantified by support, that is, the proportion of the dataset containing that item. The FP-Growth algorithm (hereinafter referred to as the FP algorithm) improves upon the AP algorithm by mining frequent itemsets and sorting them according to the number of supporting rows, constructing an FP-tree, and transforming the mining of frequent itemsets into the mining of FP-trees.

[0004] The steps of the FP algorithm are as follows:

[0005] Step 1: Input data and minimum support, filter data, generate frequent itemsets, sort them in descending order based on support, and for each frequent item, construct its conditional projection data and projected FP tree.

[0006] Step 2: Create the root node and build the FP-tree iteratively based on frequent items.

[0007] Step 3: Find the prefix path from the leaf nodes of the FP tree. If the constructed FP tree is empty, its prefix is ​​a frequent pattern; if it is not empty, use all the prefix paths to iteratively construct the FP tree.

[0008] Step 4: When there is only one path, the frequent patterns can be obtained by enumerating all possible combinations and connecting them to the prefix of this tree.

[0009] The FP algorithm constructs an FP tree by mining frequent itemsets and sorting them according to the number of supporting rows. This tree serves as the basis for determining whether there are cooperative patterns among the data. Although the algorithm solves the problem of computational complexity, it still has problems such as being unable to handle continuous data, being incompatible with different types of electronic target data, and the intermittent working characteristics of electronic targets. Summary of the Invention

[0010] This application provides a time-series-based collaborative analysis and fusion method and device for electronic targets. Using electronic target data as the object, Z-Score standardization is performed during data storage, and mean clustering is used to quantify the electromagnetic data of electronic targets. This transforms multi-feature electromagnetic data into single-feature discrete electromagnetic states, thereby better quantifying data features and significantly reducing the time complexity of collaborative computation.

[0011] This application provides a time-series-based electronic target collaborative analysis and fusion method, including:

[0012] Electromagnetic data of an electronic target is acquired, and the electromagnetic characteristics of the electromagnetic data are standardized to obtain standardized electromagnetic data.

[0013] The standardized electromagnetic data are clustered to obtain clustering results;

[0014] Quantization is performed based on the clustering results, and the quantization results are used as the discretized state data of the electronic target.

[0015] The discretized state data of the electronic target is divided into multiple sub-items;

[0016] Based on the division of multiple sub-items, the correlation coefficient between each sub-item is calculated to obtain the cooperative set of the electronic targets;

[0017] By comparing the cooperative sets of different electronic targets, the cooperative relationships between electronic targets can be determined.

[0018] Optionally, the standardization transformation of the electromagnetic characteristics of the electromagnetic data includes:

[0019] Calculate the mean and standard deviation of the electromagnetic characteristics of the electromagnetic data;

[0020] Standardization transformation is performed based on the calculated mean and standard deviation.

[0021] Optionally, clustering the standardized electromagnetic data includes:

[0022] For a standardized electromagnetic data time series, select k initial cluster centers and perform the following clustering steps:

[0023] The squared difference between the feature values ​​at each moment in the standardized electromagnetic data time series and the cluster centers of the current iteration is calculated.

[0024] The standardized electromagnetic data time series is re-divided based on the minimum squared difference to determine the clustering results for this round.

[0025] Repeat the process of determining cluster centers and performing the next iteration based on the determined cluster centers until convergence.

[0026] Optionally, quantification based on clustering results includes:

[0027] Configure the minimum and maximum number of quantizations;

[0028] The clustering steps are performed using the minimum to the maximum number of quantizations as input to obtain multiple clustering results.

[0029] Determine the silhouette coefficient of any clustering result, and select the clustering result with the largest silhouette coefficient as the quantification result of the electronic target.

[0030] Optionally, the discretized state data of the electronic target can be divided into multiple sub-items, including:

[0031] Specify the length of the sub-item;

[0032] The discretized state data of the electronic target is divided into multiple sub-items according to the sub-item length.

[0033] Optionally, based on the multiple sub-items of the partition, the correlation coefficient between each sub-item is calculated, including:

[0034] An initial correlation matrix is ​​constructed based on the partitioned sub-items. The initial correlation matrix is ​​an upper triangular matrix. Each off-diagonal sub-element in the initial correlation matrix is ​​used to record the correlation coefficient between the sub-items corresponding to the corresponding row and column. Each diagonal sub-element in the initial correlation matrix is ​​used to record the self-information entropy of the corresponding sub-item.

[0035] Configure the parameters of the initial association matrix, and then perform the following steps after configuration:

[0036] Calculate the self-information entropy of each item along the diagonal of the configured association matrix;

[0037] If the self-information entropy of any item on the diagonal is greater than the self-information threshold, calculate the correlation coefficient between each item in the corresponding row and column of the configured association matrix and the item itself.

[0038] Sub-items with correlation coefficients greater than the correlation coefficient threshold are entered into the correlation coefficient list of any given sub-item.

[0039] Optionally, comparing cooperative sets of different electronic targets to determine cooperative relationships between electronic targets includes:

[0040] Construct a collaborative set based on a list of correlation coefficients that meet the minimum sub-item number threshold;

[0041] If the cooperative sets of different electronic targets have an intersection, then it is determined that the corresponding electronic targets have a cooperative relationship.

[0042] This application also proposes a computer device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the steps of the aforementioned time-series-based electronic target collaborative analysis and fusion method.

[0043] This application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned time-series-based electronic target collaborative analysis and fusion method.

[0044] This application takes electronic target data as the object, performs Z-Score normalization transformation during data storage, and uses mean clustering to quantify the electromagnetic data of electronic targets, so that multi-feature electromagnetic data is transformed into single-feature discrete electromagnetic states, which can better quantify data features and significantly reduce the time complexity of collaborative computing.

[0045] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0046] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0047] Figure 1 This is a flowchart illustrating the electronic target collaborative analysis and fusion method according to an embodiment of this application;

[0048] Figure 2 Example of setting up sub-items for a fixed window in this application embodiment;

[0049] Figure 3 This is an example of calculating the correlation matrix in an embodiment of this application. Detailed Implementation

[0050] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0051] This application provides a time-series-based collaborative analysis and fusion method for electronic targets, such as... Figure 1 As shown, it includes:

[0052] In step S101, electromagnetic data of the electronic target is acquired, and the electromagnetic characteristics of the electromagnetic data are standardized to obtain standardized electromagnetic data. In a specific example, after acquiring the electromagnetic data of the electronic target, the electromagnetic data can be stored. In specific applications, storing electromagnetic data can adopt the following electronic target metadata model, including the following data structure:

[0053] Table 1 Data Structure of Electronic Target Metadata Model

[0054] Serial Number Parameter name Parameter type Parameter definition and content 1 uID String Electronic target unique identifier 2 uLength Int Electronic target data length 3 uPath String Electronic target data storage path 4 uType String Electronic target geographic type (sea, land, air) 5 uList JSON array Electronic target cooperative set

[0055] This data structure design enables standardized management of electromagnetic data for each target, which helps in generating collaborative discrimination data models.

[0056] In some embodiments, standardizing the electromagnetic features of the electromagnetic data includes: calculating the mean and standard deviation of the electromagnetic features of the electromagnetic data; and performing standardization based on the calculated mean and standard deviation. The Z-Score standardization transformation of the electromagnetic features (repetition frequency X', carrier frequency Y', pulse width Z') of the electronic target is performed as follows:

[0057] Using the repetition rate data of electronic target A, X' = {x1,…,x t ,…,x T For example, the processing of carrier frequency Y' and pulse width Z' is similar:

[0058] First, calculate the mean of X'. Standard deviation of X'

[0059] Calculate the standardized transformation of X' Each electromagnetic feature of the electronic target is standardized and converted, and stored according to the preset electromagnetic data model of the electronic target to form standardized electromagnetic data repetition frequency X, carrier frequency Y, and pulse width Z.

[0060] In step S102, the standardized electromagnetic data is clustered to obtain clustering results. For example, the K-means clustering method can be used for clustering and iteration.

[0061] In step S103, quantization is performed based on the clustering results, so that the quantization results are used as the discretized state data of the electronic target.

[0062] In step S104, the discretized state data of the electronic target is divided into multiple sub-items.

[0063] In step S105, based on the multiple sub-items, the correlation coefficients between each sub-item are calculated to obtain the cooperative set of the electronic targets.

[0064] In step S106, the cooperative sets of different electronic targets are compared to determine the cooperative relationships between the electronic targets.

[0065] This application takes electronic target data as the object, performs Z-Score normalization transformation during data storage, and uses mean clustering to quantify the electromagnetic data of electronic targets, so that multi-feature electromagnetic data is transformed into single-feature discrete electromagnetic states, which can better quantify data features and significantly reduce the time complexity of collaborative computing.

[0066] Clustering calculations are performed on continuous electromagnetic data to cluster electronic targets, thereby identifying and quantifying the operational states of the electronic targets. The input data for clustering is the standardized electromagnetic data of the aforementioned electronic targets; for continuous electromagnetic data, this refers to the time series of standardized electromagnetic data. In some embodiments, clustering the standardized electromagnetic data includes:

[0067] For a time series of standardized electromagnetic data, k initial cluster centers are selected, and m cluster centers are randomly selected from the time series of standardized electromagnetic data to be quantized. (0) ={m1 (0) ,…,m l (0) ,…,m k (0)} as the initial cluster centers, where Perform the following clustering steps:

[0068] The squared difference between the feature values ​​at each moment in the standardized electromagnetic data time series and the cluster centers of the current iteration is calculated. Specifically, the squared difference between the feature values ​​at each moment of the standardized electromagnetic data and m... (n) The squared difference is calculated for each class center.

[0069] The standardized electromagnetic data time series is re-divided based on the minimum squared difference to determine the clustering result for this round. For example, any feature value at any given time can be assigned to the cluster center partition with the minimum squared difference, forming a new clustering result C. (n) .

[0070] Repeat the process of determining cluster centers and performing the next iteration based on the determined cluster centers until convergence.

[0071] Specifically, the clustering result C can be analyzed. (n) Calculate the new class center and use it as the new class center m. (n+1) ={m1 (n +1) ,…,m l(n+1) ,…,m k (n+1) The new formula for calculating cluster center values ​​is as follows:

[0072]

[0073] When the iteration converges (no more new cluster centers are generated), the clustering result C is output. * If the iteration does not converge, the clustering process is repeated.

[0074] This application further quantizes the data and treats it as a discretized state of the electronic target. In some embodiments, quantization based on clustering results includes:

[0075] Configure minimum number of quantizations M min With the maximum number of quantizations M max .

[0076] From the minimum number of quantizations M min Up to the maximum number of quantizations M max The clustering steps are performed using the data as input to obtain multiple clustering results.

[0077] Determine the silhouette coefficient of any clustering result, and select the clustering result with the largest silhouette coefficient as the quantization result of the electronic target. Specifically, the silhouette coefficient can be calculated by determining clustering results with different numbers of quantizations as follows:

[0078]

[0079] Where a(k) represents the cohesion of the sample points and b(k) represents the separation.

[0080] After calculation, select the clustering result C| with the largest silhouette coefficient. Contour As the final quantization result of this electronic target, C| Contour The data is input into the electromagnetic data model of the electronic target, forming the discretized state S of the electronic target. In some specific scenarios, the electromagnetic data model of the electronic target can adopt the following data structure:

[0081] Serial Number Attribute Name Parameter type Attribute definition and content 1 Timestamp DateTime Sampling time 2 PRI Float repetition rate 3 PW Float Pulse width value 4 RF Float Carrier frequency value 5 zPRI Float Standardized repetition rate 6 zPW Float Standardized pulse width value 7 zRF Float Standardized carrier frequency value 8 Quantization Int Discretized state

[0082] This application embodiment further fixes the window setting for sub-item division. In some embodiments, dividing the discretized state data of the electronic target into multiple sub-items includes: specifying a sub-item length L, and dividing the discretized state data of the electronic target into multiple sub-items according to the sub-item length. Specifically, the sub-item length L can be set to divide the state data S of the electronic target into M sub-items of length L, such as... Figure 2 As shown, the m-th sub-item can be represented as S. m =[s Lm+1,s Lm+2 ,…,s Lm+L ].

[0083] In some embodiments, calculating the correlation coefficient between the sub-items based on the partitioning includes:

[0084] An initial correlation matrix is ​​constructed based on the partitioned sub-items. This initial correlation matrix is ​​an upper triangular matrix. Each off-diagonal element in the initial correlation matrix represents the correlation coefficient between the sub-items in the corresponding row and column, and each diagonal element represents the information entropy of the corresponding sub-item. Specifically, an initial correlation matrix Mrx is formed based on the sub-items of the electronic target. This initial correlation matrix is ​​an M×M upper triangular matrix, and each sub-element E of the initial correlation matrix... mn Represents the correlation coefficient between the m-th and n-th sub-items, where E is the element on the diagonal of the initial correlation matrix. mm H represents the self-information entropy of each sub-item.

[0085] Configure the parameters of the initial association matrix, for example, set the initial collaborative set (empty set), and in the initial association matrix, set the self-information threshold to μ∈(0,1), the correlation coefficient threshold to θ∈(0,1), and the minimum number of related sub-items σ∈Z as three parameters, and then perform the following steps after configuration:

[0086] Calculate each child element E along the diagonal of the configured association matrix. mm Self-information entropy H:

[0087]

[0088] If the self-information entropy H of any sub-element on the diagonal is greater than the self-information threshold μ, for example Figure 3 If M=5, μ=0.6, minimum number of related sub-items σ=3, and the self-information entropy H in the second row and second column is greater than the self-information threshold μ, then calculate the correlation coefficient between each sub-item in the row and column corresponding to any sub-item in the configured association matrix and the any sub-item.

[0089] Sub-items with correlation coefficients greater than the correlation coefficient threshold are entered into the correlation coefficient list of any given sub-item. Specific settings for sub-element E... mm The correlation coefficient list calculates the correlation coefficient between each item in the m-th column and the m-th row and E. mm The Pearson correlation coefficient Prs for the corresponding sub-item. If Prs > θ, then this element is entered into the correlation coefficient list. The Pearson correlation coefficient Prs can be calculated using the following steps:

[0090] Calculate the mean of the sub-items and standard deviation σ S .

[0091] Calculate the covariance between sub-items, such as sub-item s A With sub-item s B Covariance:

[0092]

[0093] Calculate the Pearson correlation coefficient between the sub-items.

[0094]

[0095] If Prs > θ, then the sub-item E is entered. mm The correlation coefficient list (or the m-th sub-item) is traversed along the diagonal, and after the traversal is completed, the collaborative set is entered into the electronic target metadata model. In some embodiments, the electronic target metadata model can adopt the following data structure:

[0096] Serial Number Parameter name Parameter type Parameter definition and content 1 uID String Electronic target unique identifier 2 uLength Int Electronic target data length 3 uPath String Electronic target data storage path 4 uType String Electronic target geographic type (sea, land, air) 5 uList JSON array Electronic target cooperative set

[0097] Here, the electronic target coherence set is a collection of all correlation coefficient lists for electronic targets, represented by a JSON array, which is a two-dimensional list. If the list of all correlation coefficients for electronic target A is [S...] t1 ,S t3 ,S t5 ],[S t2 ,S t4 ,S t6 ], where S t1 Let t1 represent the t1th sub-item. Then the cooperative set of electronic target A is [[t1,t3,t5],[t2,t4,t6]].

[0098] By designing the structure of electromagnetic data for electronic targets, the electromagnetic data of each target can be managed in a standardized manner.

[0099] In some embodiments, comparing the cooperative sets of different electronic targets to determine the cooperative relationships between electronic targets includes:

[0100] Construct a collaborative set based on a list of correlation coefficients that meet a minimum sub-item number threshold. For example... Figure 3 If the minimum number of relevant sub-items (minimum sub-item number threshold) is 3, then a cooperative set of any electronic target is constructed based on the correlation coefficient list in which the minimum number of sub-items is greater than or equal to 3.

[0101] If the cooperative sets of different electronic targets have an intersection, then the corresponding electronic targets are determined to have a cooperative relationship. For example, if the cooperative set of electronic target A is [[t1,t3,t5],[t2,t4,t6]] and the cooperative set of electronic target B is [[t1,t3,t5],[t2,t5,t8]], then the two electronic targets have the same intersection [t1,t3,t5], and it is determined that the two electronic targets have a cooperative relationship.

[0102] If multiple electronic targets have overlapping cooperative sets, they possess a cooperative relationship. These lists of electronic targets and their intersections are then entered into the electronic target cooperative model. In some examples, the electronic target cooperative model can adopt the following data structure:

[0103] Serial Number Attribute Name Parameter type Attribute definition and content 1 eID String Unique identifier for the event 2 eID_A String Electronic Target AuID 3 eID_B String Electronic Target BuID 4 uList JSON array Collaboration status

[0104] This application's method combines clustering to transform multidimensional continuous electronic target data into one-dimensional, discrete electronic target data. When calculating the association matrix, information entropy is used to filter sub-items, significantly reducing computational load. Compared to existing FP algorithms that require constructing tree structures for computation, this application's method generates association matrices and synergy sets, resulting in a simpler data structure and lower computer memory requirements.

[0105] This application also proposes a computer device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the steps of the aforementioned time-series-based electronic target collaborative analysis and fusion method.

[0106] This application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned time-series-based electronic target collaborative analysis and fusion method.

[0107] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0108] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0110] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims. All of these forms are within the protection scope of this application.

Claims

1. A time-series-based collaborative analysis and fusion method for electronic targets, characterized in that, include: Electromagnetic data of an electronic target is acquired, and the electromagnetic characteristics of the electromagnetic data are standardized to obtain standardized electromagnetic data. Configure the minimum and maximum number of quantizations; Using the minimum to maximum number of quantizations as input, a clustering step is performed to cluster the standardized electromagnetic data and obtain multiple clustering results; Determine the silhouette coefficient of any clustering result, and select the clustering result with the largest silhouette coefficient as the quantification result of the electronic target; The quantization results are used as the discretized state data of the electronic target; The discretized state data of the electronic target is divided into multiple sub-items; Based on the division of multiple sub-items, the correlation coefficient between each sub-item is calculated to obtain the cooperative set of the electronic targets; By comparing the cooperative sets of different electronic targets, the cooperative relationships between electronic targets can be determined.

2. The time-series-based electronic target collaborative analysis and fusion method as described in claim 1, characterized in that, The standardization transformation of the electromagnetic characteristics of the electromagnetic data includes: Calculate the mean and standard deviation of the electromagnetic characteristics of the electromagnetic data; Standardization transformation is performed based on the calculated mean and standard deviation.

3. The time-series-based electronic target collaborative analysis and fusion method as described in claim 1, characterized in that, Clustering the standardized electromagnetic data includes: For time series of standardized electromagnetic data, select k initial cluster centers and perform the following clustering steps: The squared difference between the feature values ​​at each moment in the time series of the standardized electromagnetic data and the cluster centers of the current iteration is calculated. The time series of the standardized electromagnetic data is re-divided based on the minimum squared difference to determine the clustering results for this round. Repeat the process of determining cluster centers and performing the next iteration based on the determined cluster centers until convergence.

4. The time-series-based electronic target collaborative analysis and fusion method as described in claim 3, characterized in that, The discretized state data of the electronic target is divided into multiple sub-items, including: Specify the length of the sub-item; The discretized state data of the electronic target is divided into multiple sub-items according to the sub-item length.

5. The time-series-based electronic target collaborative analysis and fusion method as described in claim 4, characterized in that, Based on the division into multiple sub-items, the correlation coefficient between each sub-item is calculated as follows: An initial correlation matrix is ​​constructed based on the partitioned sub-items. The initial correlation matrix is ​​an upper triangular matrix. Each off-diagonal sub-element in the initial correlation matrix is ​​used to record the correlation coefficient between the sub-items corresponding to the corresponding row and column. Each diagonal sub-element in the initial correlation matrix is ​​used to record the self-information entropy of the corresponding sub-item. Configure the parameters of the initial association matrix, and then perform the following steps after configuration: Calculate the self-information entropy of each item along the diagonal of the configured association matrix; If the self-information entropy of any item on the diagonal is greater than the self-information threshold, calculate the correlation coefficient between each item in the corresponding row and column of the configured association matrix and the item itself. Sub-items with correlation coefficients greater than the correlation coefficient threshold are entered into the correlation coefficient list of any sub-element.

6. The time-series-based electronic target collaborative analysis and fusion method as described in claim 5, characterized in that, Comparing the cooperative sets of different electronic targets to determine the cooperative relationships between them includes: Construct a collaborative set based on a list of correlation coefficients that meet the minimum sub-item number threshold; If the cooperative sets of different electronic targets have an intersection, then it is determined that the corresponding electronic targets have a cooperative relationship.

7. A computer device, characterized in that, It includes a processor and a memory, the memory storing a computer program that, when executed by the processor, implements the steps of the time-series-based electronic target collaborative analysis and fusion method as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the time-series-based electronic target collaborative analysis and fusion method as described in any one of claims 1 to 6.

Images