A radar cooperative detection strategy generation method based on a soft clustering algorithm

Abstract

The application provides a cooperative detection strategy generation method based on a soft clustering algorithm, which can combine a cooperative node topological structure and common link states and special link states among nodes, adopt a K-means soft clustering algorithm to quickly and effectively generate task group information, adopt a scoring system to complete center node selection and confirmation and group master node selection and confirmation, and select a task execution node by the group master node, fully considers the aggregation degree among the cooperative nodes and the importance of the roles of the center node and the group master node, can provide a decision-making suggestion for implementation of multi-node cooperative detection, and has important guarantee significance for the cooperative detection.

Description

Technical Field

[0001] This invention belongs to the field of radar mission scheduling technology. Background Technology

[0002] With the increasing complexity and variability of target detection environments, single-platform detection can no longer meet practical needs, and multi-node collaboration has become a trend. Collaborative nodes can perform various tasks such as collaborative search, collaborative tracking, collaborative localization, collaborative anti-interference, and guided detection. In particular, the rational organization and planning of collaborative detection tasks among multiple nodes is crucial for improving collaborative detection capabilities. In a multi-node detection system, the roles of nodes are generally assigned according to the task, and different tasks are performed based on these roles. Roles have primary and secondary roles, and their corresponding task decision-making and execution authority differ.

[0003] The patents "A Radar Formation Cooperative Detection Method Based on a Multi-Task Orderly Traction Strategy" and "Control System and Method for UAV Formation Cooperative Flight Based on Edge Cloud" only allocate tasks and schedule resources for all nodes within a large formation, without mentioning task group division. Node roles are only indicated as master and slave nodes, without further detailed classification. However, in cooperative detection system mission execution, tasks are often organized and planned within a large cooperative network, lacking task group division. Appropriate task group division can improve both cooperative mission execution efficiency and cooperative detection efficiency. However, the criteria for task group division vary, leading to inconsistent system robustness. In particular, the different choices of central nodes, group master nodes, and task nodes have a significant impact on the implementation of cooperative detection. Therefore, how to rationally assign node roles has a substantial influence on the actual performance of cooperative detection missions. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention proposes a radar cooperative detection strategy generation method based on a soft clustering algorithm, thereby achieving multi-level cooperative detection between nodes within a group and between task groups. The cooperative detection strategy includes task center node selection, task group division, task group master node selection, and task node designation. The task center node is the highest priority and decision-making authority among all nodes in the cooperative network. It is responsible for task planning and decomposition based on cooperative detection task requests and the cooperative node topology, generating task groups and task group subtasks. The task group master node is the highest priority and decision-making authority within a task group. It is responsible for selecting task nodes for the cooperative detection task group subtasks and generating, evaluating, and adjusting the cooperative detection scheme for the group based on the cooperative detection subtasks. Task nodes are the specific execution nodes of the cooperative detection scheme. Task nodes belong to a certain task group, and there is no task overlap between nodes in different task groups. The center node and the group master node can also be task nodes.

[0005] The present invention proposes a radar cooperative detection strategy generation method based on a soft clustering algorithm, the technical solution of which is as follows:

[0006] Step 1: Collaborative nodes join the network;

[0007] Step 2: Central Node Election. Calculate the score of the central node elected by each collaborative node in the collaborative network.

[0008] Step 3: Confirm the central node. For the set of central nodes that have been selected, choose the node with the highest score as the central node C. cn , where cn is the central node number, and the confirmed central node information is pushed to other nodes in the collaborative network through ordinary links;

[0009] Step 4: Determine the number K of task groups;

[0010] Step 5: Use the K-means clustering algorithm to generate K task groups;

[0011] Step 6: Group leader node election, calculate the score of the group leader node elected by each collaborative node in the task group;

[0012] Step 7: Confirm the group leader node. For the group of group leader nodes that have been elected, select the node with the highest score as the group leader node G. gn gn is the group master node number, and the confirmed group master node information is pushed to other nodes and the central node in the group through the ordinary link.

[0013] Step 8: Select task nodes. Choose a combination of nodes that can complete the collaborative exploration sub-tasks as task execution nodes.

[0014] The beneficial effects of this invention are as follows:

[0015] This invention proposes a radar cooperative detection strategy generation method based on a soft clustering algorithm. It combines the cooperative node topology and the ordinary and special link states between nodes, employing a K-means soft clustering algorithm to quickly and effectively generate task group information. A scoring system is used to select and confirm the central node and the group leader node, which then selects the task execution node. This method fully considers the aggregation degree between cooperative nodes and the importance of the roles of the central node and the group leader node, providing a decision-making suggestion for the implementation of multi-node cooperative detection and offering significant assurance for cooperative detection. Attached Figure Description

[0016] Figure 1 This is a preferred flowchart of the present invention. Detailed Implementation

[0017] The invention will be illustrated with reference to the accompanying drawings.

[0018] The present invention proposes a radar cooperative detection strategy generation method based on soft clustering algorithm, and the preferred implementation process is shown in the appendix. Figure 1 Specifically, the process is described as follows:

[0019] Step 1: Collaborating nodes join the network. Let the number of collaborating nodes in the network be N, and the link status between the collaborating nodes be nl. i,j (s c ,w c ,s s ,w s ) including ordinary links c and special links s They share their location information gl(n,lon,lat,h) via ordinary links, where n is the cooperating node number, lon is the node longitude, lat is the node latitude, h is the node height, and s is the node height. c For the valid state of the normal link between the i-th node and the j-th node, s c =1 indicates that it is valid, s c =0 indicates invalid, w c This represents the transmission bandwidth between two nodes when a normal link is active; s s For the specific effective state of the link between the i-th node and the j-th node, s s =1 indicates that it is valid, s s =0 indicates invalid, w s This refers to the transmission bandwidth between two nodes when a special link is active.

[0020] Step 2: Central Node Election. Calculate the score of the central node elected by each collaborative node in the collaborative network.

[0021] The optimization process is as follows: Let the node number be k n , 1≤n≤N, the kth n The central node selected by each node is numbered i. n Calculate the i-th n The sum of the distances between each node and other nodes in the collaborative network Get the i-th n Distance state factor of each node Where r(i n ,j) is the i-th n latitude and longitude of each node gl(i) n The distance between two nodes calculated from gl(j,lon,lat,h) and the j-th node gl(j,lon,lat,h); calculate the distance between the i-th node and the j-th node gl(j,lon,lat,h); n The sum of the valid state values ​​of ordinary links between each node and other nodes Get the i-th nSpecial link state factor between each node and other nodes Calculate the i-th n The state score C of each node as the central node c (i n )=λ1R e (i n )+λ2S e (i n ), where λ1 and λ2 are weighting coefficients, and λ1 + λ2 = 1;

[0022] Step 3: Confirm the central node. For the set of central nodes that have been selected, choose the node with the highest score as the central node C. cn , where cn is the central node number, and the confirmed central node information is pushed to other nodes in the collaborative network through ordinary links;

[0023] Step 4: Determine the number of task groups K, where K ranges from 1 to N. The preferred method is as follows:

[0024] Step 4-1: For each K value, execute the K-means clustering algorithm;

[0025] Step 4-1-1: Initialize K cluster centers;

[0026] Step 4-1-2: Distribute the central node to each group;

[0027] Step 4-1-3: Assign nodes to the nearest cluster center;

[0028] Step 4-1-4: Update cluster centers;

[0029] Step 4-1-5: Repeat the above steps until convergence.

[0030] Step 4-2: For each K value, calculate the sum of the squared distances.

[0031] Step 4-2-1: Calculate the sum of squared distances from each node in each cluster to its cluster center;

[0032] Step 4-2-2: Sum the squared distances of each group to obtain the total squared distance ΓR. 2 (k),k∈[1,N].

[0033] Step 4-3, for a set of ΓR 2 (k), to find the optimal value of K;

[0034] Step 4-3-1: For k∈[2,N-1], calculate the rate of change of slope between its two adjacent points;

[0035] Step 4-3-2: Select the k value corresponding to the point with the largest rate of change of slope as the optimal K value.

[0036] Step 5: Generate K task groups using the K-means clustering algorithm. The specific method is as follows:

[0037] Step 5-1: Initialize K cluster centers;

[0038] Step 5-2: Distribute the central node to each group;

[0039] Step 5-3: Assign nodes to the nearest cluster center;

[0040] Step 5-4: Update cluster centers;

[0041] Step 5-5: Repeat the above steps until convergence.

[0042] Step 6: Group leader node election, calculate the score of the group leader node elected by each collaborative node in the task group;

[0043] The preferred method is as follows: Given that the center node is C cn The k-th task group (1≤k≤K) has M nodes, and the nodes within the group are numbered k. m , 1≤m≤M, let the kth m The group leader node elected by each node is numbered i. m Calculate the i-th m The sum of the distances between each node and other nodes in the task group Get the i-th m Distance state factor of each node Where r(i m ,j) is the i-th m latitude and longitude of each node gl(i) m The distance between two nodes calculated from gl(j,lon,lat,h) and the j-th node gl(j,lon,lat,h); calculate the distance between the i-th node and the j-th node gl(j,lon,lat,h); m The sum of the valid state values ​​of the special links between each node and other nodes Get the i-th m Special link state factor between each node and other nodes Calculate the i-th m The state score of each node as the group owner node Where s s (i m ,cn) is the i-th m The special link states between each node and the central node, η1 and η2 are weighting coefficients, and η1+η2=1.

[0044] Step 7: Confirm the group leader node. For the group of group leader nodes that have been elected, select the node with the highest score as the group leader node G. gn gn is the group master node number, and the confirmed group master node information is pushed to other nodes and the central node in the group through the ordinary link.

[0045] Step 8: Task Node Selection. The group master node selects a combination of nodes capable of completing the collaborative detection sub-tasks based on the collaborative detection scheme and the link status and topology relationships between nodes in the group. These nodes serve as task execution nodes. The methods for determining whether a node can complete a collaborative detection sub-task include, but are not limited to, whether the node's detection power covers the detection target and whether the node's detection data rate meets the task requirements.

[0046] This invention fully considers the aggregation degree between collaborative nodes and the importance of the roles of the central node and the group master node, and can provide a decision-making suggestion for the implementation of multi-node collaborative detection, which has important guarantee significance for collaborative detection.

Claims

1. A radar cooperative detection strategy generation method based on soft clustering algorithm, characterized in that: Step 1: Connect the cooperative radar nodes to the network. Let the number of cooperative radar nodes in the network be N, and the link status between the cooperative radar nodes be nl. i,j (s c ,w c ,s s ,w s ) including ordinary links c and special links s They share their location information gl(n,lon,lat,h) via ordinary links, where n is the cooperative radar node number, lon is the longitude of the radar node, lat is the latitude of the radar node, h is the altitude of the radar node, and s is the position of the radar node. c For the effective state of the normal link between the i-th radar node and the j-th radar node, s c =1 indicates that it is valid, s c =0 indicates invalid, w c This represents the transmission bandwidth between two radar nodes when a normal link is active; s s For the specific effective state of the link between the i-th radar node and the j-th radar node, s s =1 indicates that it is valid, s s =0 indicates invalid, w s This refers to the transmission bandwidth between two radar nodes when a special link is active. Step 2: Selection of the central radar node; calculate the score of the central radar node selected by each cooperative radar node in the cooperative network. Step 3: Confirm the central radar node. For the group of central radar nodes that have been selected, choose the radar node with the highest score as the central radar node C. cn , where cn is the central radar node number, and the confirmed central radar node information is pushed to other radar nodes in the collaborative network through ordinary links; Step 4: Determine the number K of task groups; Step 5: Use the K-means clustering algorithm to generate K task groups; Step 6: Group leader radar node selection, calculate the score of the group leader radar node selected by each cooperating radar node in the task group; Step 7: Confirm the group leader radar node. For the group of group leader radar nodes that have been selected, choose the radar node with the highest score as the group leader radar node G. gn gn is the group leader radar node number, and the confirmed group leader radar node information is pushed to other radar nodes and the central radar node in the group through a normal link. Step 8: Select the radar node combination that can complete the collaborative detection sub-task as the radar node for mission execution.

2. The radar cooperative detection strategy generation method based on soft clustering algorithm according to claim 1, characterized in that: Step 2 also includes: assigning the radar node number k. n , 1≤n≤N, the kth n The central radar node selected by the radar nodes is numbered i. n Calculate the i-th n The sum of the distances between each radar node and other radar nodes in the cooperative network Where r(i n ,j) is the i-th n The distance between the i-th radar node and the j-th radar node is obtained. n Range state factor of each radar node Calculate the i-th n The sum of the valid state values ​​of the ordinary links between each radar node and other radar nodes Get the i-th n Common link state factors between each radar node and other radar nodes Calculate the i-th n The state score C of each radar node as the central radar node c (i n )=λ1R e (i n )+λ2S e (i n ), where λ1 and λ2 are weighting coefficients, and λ1+λ2=1.

3. The radar cooperative detection strategy generation method based on soft clustering algorithm according to claim 1, characterized in that: In step 4, the number of task groups K is determined, with K ranging from 1 to N, including: Step 4-1: For each K value, execute the K-means clustering algorithm; Step 4-1-1: Initialize K cluster centers; Step 4-1-2: Assign the central radar node to each group; Step 4-1-3: Assign radar nodes to the nearest cluster center; Step 4-1-4: Update cluster centers; Step 4-1-5: Repeat the above steps until convergence. Step 4-2: For each K value, calculate the sum of the squared distances. Step 4-2-1: Calculate the sum of squared distances from each radar node in each group to its cluster center; Step 4-2-2: Sum the squared distances of each group to obtain the total squared distance ΓR. 2 (k), k∈[1,N]; Step 4-3, for a set of ΓR 2 (k), to find the optimal value of K; Step 4-3-1: For k∈[2,N-1], calculate the rate of change of slope between its two adjacent points; Step 4-3-2: Select the k value corresponding to the point with the largest rate of change of slope as the optimal K value.

4. The radar cooperative detection strategy generation method based on soft clustering algorithm according to claim 1, characterized in that: Step 6 further includes: The central radar node is known to be C. cn The number of radar nodes in the k-th task group (1≤k≤K) is M, and the radar node numbers within the group are k. m , 1≤m≤M, let the kth m The group leader radar node selected by the radar nodes is numbered i. m Calculate the i-th m The sum of the distances between each radar node and other radar nodes within the mission group Where r(i m ,j) is the i-th m The distance between the i-th radar node and the j-th radar node is obtained. m Range state factor of each radar node Calculate the i-th m The sum of the valid state values ​​of the special links between each radar node and other radar nodes Get the i-th m Special link state factor between each radar node and other radar nodes Calculate the i-th m The state score of each radar node as the group leader radar node Where s s (i m ,cn) is the i-th m The special link states between each radar node and the central radar node are η1 and η2, which are weighting coefficients, and η1 + η2 = 1.

Images