Unmanned aerial vehicle network cluster maintenance method

Through dynamic management of drone network cluster merging, splitting and cluster head replacement, the problem of insufficient energy consumption and topological adaptability of multi-channel system networks in the prior art is solved, and the performance of drone hierarchical clustering network is realized Improve and optimize energy consumption.

CN119012303APending Publication Date: 2024-11-22BEIJING SENTRY TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410873075.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing clustering algorithm for hierarchical clustering networks fails to effectively consider the network energy consumption of multi-channel systems, resulting in insufficient network performance, especially in terms of low power consumption and topological adaptation requirements.

Method used

A method for maintaining a drone network cluster is proposed. Through dynamic management of cluster merging, cluster splitting and cluster head replacement, the cluster structure is optimized by using inter-cluster merger values ​​and cluster splitting values, and the appropriate cluster heads are selected to improve network performance. The specific steps include calculating the inter-cluster merge value when cluster merging, selecting the smallest cluster as the cluster to be merged, calculating the cluster split value when cluster splitting, selecting the smallest node as the new cluster head, and according to the node weight when cluster heads are replaced Make replacement.

Benefits of technology

By dynamically managing cluster structure and cluster heads, the network performance of the drone layered clustered network is improved, the network survival time is extended, energy consumption is reduced, and the needs of multi-channel systems are adapted.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119012303A_ABST
    Figure CN119012303A_ABST
Patent Text Reader

Abstract

The invention provides an unmanned aerial vehicle network cluster maintenance method, which comprises the following steps that: in response to a cluster head of any cluster of an unmanned aerial vehicle network, other clusters of which the total number of nodes is not greater than a preset number of nodes in the cluster after being merged with any cluster exist around are known, and the cluster head of any cluster determines a to-be-merged cluster from the other clusters; the cluster head of any cluster sends cluster merging request information to the cluster head of the to-be-merged cluster; after the cluster head of the to-be-merged cluster receives the cluster merging request information, if the cluster merging is agreed, cluster merging agreement response information is sent to the node in any cluster; and after the node in any cluster receives the cluster merging agreement response information, changing the cluster head into the cluster head of the cluster to be merged. According to the unmanned aerial vehicle network cluster maintenance method provided by the invention, the problem that the network survival time is relatively short in the cluster maintenance stage is solved, and the network performance of the unmanned aerial vehicle layered and clustered network is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer and communication technology, and in particular to a method for maintaining a drone network cluster. Background Art

[0002] UAV networks play an important role in disaster relief and other scenarios. Currently, FANET (Flying Ad-Hoc Network) is mainly divided into two types: parallel structure and hierarchical clustering structure. The hierarchical clustering structure divides FANET into clusters of different sizes, and manages the clusters uniformly through cluster head nodes, which can not only reduce the amount of message interaction in network management, but also reduce the overhead of flooding routing, and can well adapt to the requirements of stability and robustness in UAV disaster relief scenarios. At present, clustering algorithms for multi-channel systems are mostly based on spectrum sensing to allocate channels between clusters, or assist cluster head selection based on the number of channels available to each node, but do not consider the impact of clustering algorithms on the network energy consumption of multi-channel systems. In the face of the requirements of UAV networks for low power consumption and topology adaptation, it is necessary to propose a cluster maintenance method that can improve the network performance of UAV hierarchical clustering networks. Summary of the invention

[0003] The present invention provides a UAV network cluster maintenance method to solve the defect that the clustering algorithm in the prior art does not consider the impact on the network energy consumption of the multi-channel system, and realizes a cluster maintenance method that can improve the network performance of the UAV hierarchical clustering network.

[0004] The present invention provides a method for maintaining a drone network cluster, comprising the following steps: in response to a cluster head of any cluster of the drone network learning that there are other clusters around the cluster whose total number of nodes after merging with the any cluster does not exceed the number of nodes in a preset cluster, the cluster head of any cluster determines a cluster to be merged from the other clusters; the cluster head of any cluster sends a cluster merging request message to the cluster head of the cluster to be merged; after the cluster head of the cluster to be merged receives the cluster merging request message, if the cluster head agrees to the cluster merging, the cluster head sends a cluster merging agreement response message to the nodes in any cluster; after the nodes in any cluster receive the cluster merging agreement response message, the cluster head is changed to the cluster head of the cluster to be merged.

[0005] According to a method for maintaining a drone network cluster provided by the present invention, the cluster head of any cluster determines a cluster to be merged from the other clusters, including: the cluster head of any cluster calculates an inter-cluster merging value with the other clusters, and selects a cluster with the smallest inter-cluster merging value as the cluster to be merged; the inter-cluster merging value is expressed as:

[0006] in, Represents clustering and clustering The inter-cluster merge value between, Represents clustering and clustering The number of nodes with communication links in represents the number of nodes in the preset cluster, Represents clustering The number of nodes, Represents clustering The number of nodes, Represents clustering Cluster Head and clustering The distance between cluster heads, , , Represents the weighting coefficient.

[0007] According to a drone network cluster maintenance method provided by the present invention, the method also includes: after the cluster head of the cluster to be merged receives the cluster merge request information, if it does not agree to the cluster merge, it sends a cluster merge rejection response information to the nodes in any of the clusters; after the cluster head of any of the clusters receives the cluster merge rejection response information, it redefines the cluster to be merged in the other clusters until it receives the cluster merge agreement response information sent by the redetermined cluster head of the cluster to be merged; wherein the redetermined inter-cluster merge value of the cluster to be merged and the any cluster is second only to the previously determined inter-cluster merge value of the cluster to be merged and the any cluster.

[0008] According to a method for maintaining a drone network cluster provided by the present invention, the method also includes: in response to the cluster head of any sub-cluster of the drone network learning that the number of nodes in the any sub-cluster exceeds a preset node number threshold, the cluster head of any sub-cluster determines a newly added cluster head for cluster splitting from the any sub-cluster; the cluster head of any sub-cluster broadcasts cluster splitting prompt information to other nodes in the any sub-cluster; wherein the cluster splitting prompt information includes information of the newly added cluster head for cluster splitting; after the newly added cluster head for cluster splitting receives the cluster splitting prompt information, if it agrees to cluster splitting, it broadcasts the information of agreeing to cluster splitting to other nodes in the any sub-cluster; after the nodes in any sub-cluster receive the information of agreeing to cluster splitting, they determine whether the cluster head is updated to the newly added cluster head for cluster splitting by calculating the cluster splitting value, and if it needs to be updated to the newly added cluster head for cluster splitting, the cluster head information is updated.

[0009] According to a drone network cluster maintenance method provided by the present invention, the cluster splitting value is expressed as:

[0010] in, Representation Node and nodes The cluster splitting value, Representation Node and nodes The distance Representation Node and nodes The speed difference, , Represents the weighting coefficient.

[0011] According to a drone network cluster maintenance method provided by the present invention, the cluster head of any sub-cluster determines a new cluster head for cluster splitting from any sub-cluster, including: the cluster head of any sub-cluster calculates the node weights of other nodes in any sub-cluster except itself, and selects the node with the smallest node weight as the new cluster head for cluster splitting.

[0012] According to a drone network cluster maintenance method provided by the present invention, the node weight is expressed as:

[0013]

[0014] in, Representation Node The node weights of Representation Node The node difference of Representation Node The node degree, represents the number of nodes in the preset cluster, Representation Node The distance to neighbor nodes and Representation Node The relative moving speed with neighboring nodes and, Representation Node Power consumed, , , , Represents the weighting coefficient.

[0015] According to a method for maintaining a drone network cluster provided by the present invention, the method further includes: after the newly added cluster head of the cluster splitting receives the cluster splitting prompt information, if it does not agree with the cluster splitting, it broadcasts a cluster splitting rejection information to the nodes in any of the sub-clusters; after the cluster head of any of the sub-clusters receives the cluster splitting rejection information, it redetermines the newly added cluster head of the cluster splitting from any of the sub-clusters until it receives the cluster splitting agreement information sent by the redetermined newly added cluster head of the cluster splitting; wherein the node weight of the newly determined newly added cluster head of the cluster splitting is second only to the node weight of the previously determined newly added cluster head of the cluster splitting.

[0016] According to a drone network cluster maintenance method provided by the present invention, the method also includes: in response to the cluster head of any sub-cluster of the drone network learning that there is a node in any sub-cluster whose node weight is smaller than itself, determining the node with the smallest node weight as the cluster head to be replaced; the cluster head of any sub-cluster sends a cluster head replacement request to the cluster head to be replaced; after the cluster head to be replaced receives the cluster head replacement request, if it agrees to become the new cluster head, it broadcasts a message of consent to update the cluster head to other nodes in any sub-cluster; after the nodes in any sub-cluster receive the message of consent to update the cluster head, they update the cluster head to the cluster head to be replaced.

[0017] According to a drone network cluster maintenance method provided by the present invention, the method also includes: after the cluster head to be replaced receives the cluster head replacement request, if the cluster head to be replaced does not agree to become the new cluster head, it broadcasts a cluster head update refusal message to other nodes in any of the sub-clusters; after the cluster head of any of the sub-clusters receives the cluster head update refusal message, if it is learned that there are still nodes with node weights smaller than itself in any of the sub-clusters, it determines the node with the smallest node weight among the nodes that have not been selected as the cluster head to be replaced as the new cluster head to be replaced, until the cluster head agrees to update the cluster head message is received.

[0018] The unmanned aerial vehicle network cluster maintenance method provided by the present invention learns, in response to the cluster head of any cluster of the unmanned aerial vehicle network, that there are other clusters around whose total number of nodes after merging with any cluster does not exceed the number of nodes in a preset cluster, the cluster head of any cluster determines the cluster to be merged from the other clusters, the cluster head of any cluster sends cluster merging request information to the cluster head of the cluster to be merged, after the cluster head of the cluster to be merged receives the cluster merging request information, if it agrees to the cluster merging, it sends cluster merging agreement response information to the nodes in any cluster, after receiving the cluster merging agreement response information, the nodes in any cluster change the cluster head to the cluster head of the cluster to be merged, thereby realizing the merging of smaller clusters, solving the problem of short network survival time in the cluster maintenance stage, and improving the network performance of the unmanned aerial vehicle hierarchical cluster network. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or 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 creative work.

[0020] Figure 1 This is one of the flow diagrams of cluster merging in the drone network cluster maintenance method provided by the present invention.

[0021] Figure 2 This is the second flow chart of cluster merging in the UAV network cluster maintenance method provided by the present invention.

[0022] Figure 3 This is one of the flow charts of cluster splitting in the UAV network cluster maintenance method provided by the present invention.

[0023] Figure 4 This is the second flow chart of cluster splitting in the UAV network cluster maintenance method provided by the present invention.

[0024] Figure 5 This is one of the flow charts of cluster head replacement in the UAV network cluster maintenance method provided by the present invention.

[0025] Figure 6 This is the second flow chart of cluster head replacement in the UAV network cluster maintenance method provided by the present invention.

[0026] Figure 7 It is a comparison chart of the average residual energy consumption of all network nodes in the dual-channel system between the LP-WCA algorithm and the WCA algorithm.

[0027] Figure 8 It is a schematic diagram of the simulation results of the number of clusters changing over time when the LP-WCA algorithm and the WCA algorithm are running in a dual-channel system.

[0028] Fig. 9 The figure is a comparison of the network task cycles of the LP-WCA algorithm and the WCA algorithm in a dual-channel system. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Among the existing clustering algorithms, the weighted clustering algorithm WCA can comprehensively consider factors such as node mobility, topological connection, location and energy consumption for clustering. The cluster head selected by the algorithm will have advantages in terms of geographical location, node connectivity and energy consumption.

[0031] When the WCA algorithm selects a cluster head, it first weights the four factors of node difference, distance between the node and its neighbors, relative moving speed between the node and its neighbors, and energy consumed by the node to obtain the node weight. Then, it compares the node weights to select the node with the smallest weight as the cluster head. The calculation formula is as follows:

[0032] In the formula, Representation Node The node weights of , , , represents the weighting coefficient, and + + + =1. The smaller the value, the better the performance of the node, and the more suitable it is to be the cluster head node in the current area.

[0033] In the formula, Representation Node The node difference of , Representation Node The node degree, Indicates the preset number of nodes in the cluster, which represents the optimal number of nodes in the cluster. Node difference is used to evaluate the size of the node's neighbors. The smaller the node difference, the more appropriate the number of neighbors of the node. If it becomes the cluster head, the number of nodes in the cluster is closer to the optimal size.

[0034] In the formula, Representation Node The sum of the distances to neighboring nodes represents the distance between a node and other nodes, indicating the "geographic location" of a node. The smaller the sum of the distances to neighboring nodes, the smaller the node is at the relative center of the geographic location between itself and its neighboring nodes.

[0035] In the formula, Representation Node The relative moving speed and relative moving speed of the node with neighboring nodes represent whether the node can maintain link connectivity with neighboring nodes for a long time, thereby achieving long-term communication. The smaller the relative moving speed and relative moving speed of the node, the longer the node can maintain relative position with surrounding nodes.

[0036] In the formula, Representation Node The consumed power represents the energy consumed by the node in its previous work.

[0037] The present invention proposes an improved weighted clustering algorithm LP-WCA (Low Power Weighted Cluster Algorithm) based on the WCA algorithm for unmanned aerial vehicle ad hoc networks to improve the network energy consumption and robustness performance of unmanned aerial vehicle networks.

[0038] Figure 1 This is one of the flow diagrams of cluster merging in the UAV network cluster maintenance method provided by the present invention. Figure 1 As shown in the figure, the cluster merging process includes: Step 101: In response to the cluster head of any cluster of the drone network learning that there are other clusters around the cluster whose total number of nodes after merging with the any cluster does not exceed the number of nodes in the preset cluster, the cluster head of any cluster determines a cluster to be merged from the other clusters.

[0039] The preset number of nodes in the cluster represents the optimal number of nodes in the cluster. If the cluster head of any cluster in the UAV network learns that there are other clusters around whose total number of nodes after merging with any cluster does not exceed the preset number of nodes in the cluster, the cluster head of any cluster determines the cluster to be merged from other clusters.

[0040] Step 102: The cluster head of any cluster sends a cluster merging request message to the cluster head of the cluster to be merged.

[0041] The cluster head of any sub-cluster sends cluster merging request information to the cluster head of the sub-cluster to be merged, so as to request the cluster head of the sub-cluster to be merged to merge with the sub-cluster to be merged.

[0042] Step 103: After receiving the cluster merging request information, if the cluster head of the sub-cluster to be merged agrees to the cluster merging, it sends a cluster merging agreement response information to the nodes in any of the sub-clusters.

[0043] After receiving the cluster merge request information, if the cluster head of the cluster to be merged agrees to the cluster merge, it sends a cluster merge agreement response information to the nodes in any cluster. The cluster head of the cluster to be merged can send the cluster merge agreement response information to each node in any cluster by broadcasting. The cluster merge agreement response information may include the information of the cluster head of the cluster to be merged and the indication information of agreeing to the cluster merge.

[0044] Step 104: After receiving the cluster merging agreement response information, the node in any of the sub-clusters changes the cluster head to the cluster head of the sub-cluster to be merged.

[0045] After receiving the cluster merging agreement response information, the node in any sub-cluster changes the cluster head to the cluster head of the sub-cluster to be merged, so that any sub-cluster and the sub-cluster to be merged are merged into one cluster, and the cluster head is changed to the cluster head of the determined sub-cluster to be merged.

[0046] The unmanned aerial vehicle network cluster maintenance method provided by the present invention learns, in response to the cluster head of any cluster of the unmanned aerial vehicle network, that there are other clusters around whose total number of nodes after merging with any cluster does not exceed the number of nodes in a preset cluster, the cluster head of any cluster determines the cluster to be merged from the other clusters, the cluster head of any cluster sends cluster merging request information to the cluster head of the cluster to be merged, after the cluster head of the cluster to be merged receives the cluster merging request information, if it agrees to the cluster merging, it sends cluster merging agreement response information to the nodes in any cluster, after receiving the cluster merging agreement response information, the nodes in any cluster change the cluster head to the cluster head of the cluster to be merged, thereby realizing the merging of smaller clusters, solving the problem of short network survival time in the cluster maintenance stage, and improving the network performance of the unmanned aerial vehicle hierarchical cluster network.

[0047] According to a method for maintaining a drone network cluster provided by the present invention, the cluster head of any cluster determines a cluster to be merged from the other clusters, including: the cluster head of any cluster calculates an inter-cluster merging value with the other clusters, and selects a cluster with the smallest inter-cluster merging value as the cluster to be merged; the inter-cluster merging value is expressed as:

[0048] in, Represents clustering and clustering The inter-cluster merge value between, Represents clustering and clustering The number of nodes with communication links in represents the number of nodes in the preset cluster, Represents clustering The number of nodes, Represents clustering The number of nodes, Represents clustering Cluster Head and clustering The distance between cluster heads, , , Represents the weighting coefficient.

[0049] When the cluster head of any cluster determines the cluster to be merged from other clusters, it calculates the inter-cluster merging values ​​with other clusters and selects the cluster with the smallest inter-cluster merging value as the cluster to be merged.

[0050] The inter-cluster merging value is calculated by weighted calculation. In the expression of the inter-cluster merging value, Represents clustering and clustering The number of nodes (neighboring nodes) with communication links in the cluster determines the degree of interleaving between the two clusters. Indicates the number of nodes in the preset cluster. Represents clustering The number of nodes, Represents clustering The number of nodes, Represents whether the size of the merged cluster is close to the optimal size. Represents clustering Cluster Head and clustering The distance between the cluster heads determines the range of the merged cluster. The bigger, The smaller it is, the more adjacent nodes there are between the two clusters, that is, the greater the degree of interweaving; The smaller it is, the closer the two clusters are to the maximum size of the cluster after merging; The smaller the value, the closer the two cluster heads are. The smaller it is, the more suitable the two clusters are for merging. , , represents the weighting coefficient, + + =1. You can set = = =0.33.

[0051] Figure 2 This is the second flow chart of cluster merging in the UAV network cluster maintenance method provided by the present invention. Figure 2 The process of cluster merging is given through a specific example. Figure 2 As shown, the method includes: Step 1: When the cluster Cluster Head When other clusters are found around and the size of the merged cluster will not exceed the preset maximum value, The merging process with the neighboring cluster will be started. When there are multiple clusters around that meet the merging scale, the cluster head The calculation with each neighboring cluster value (inter-cluster merge value), and then select The smallest neighboring clusters are merged. Calculated After the minimum, the cluster head Towards the neighboring cluster head send This message indicates that the cluster head is The cluster preparation and cluster head are The clusters are merged.

[0052] Step 2: When the cluster head v6 node receives After the message, All nodes publish Message, indicating cluster Agree with cluster merge.

[0053] Step 3: When the cluster middle Node receives After the message, send Message, indicating Has received the merge message and officially joined the cluster head The cluster where and The same goes for nodes. At this point, the two clusters are officially merged into a new cluster.

[0054] The UAV network cluster maintenance method provided by the present invention calculates the inter-cluster merging value with other clusters and selects the cluster with the smallest inter-cluster merging value as the cluster to be merged, thereby improving the rationality of determining the clusters to be merged.

[0055] According to a drone network cluster maintenance method provided by the present invention, the method also includes: after the cluster head of the cluster to be merged receives the cluster merge request information, if it does not agree to the cluster merge, it sends a cluster merge rejection response information to the nodes in any of the clusters; after the cluster head of any of the clusters receives the cluster merge rejection response information, it redefines the cluster to be merged in the other clusters until it receives the cluster merge agreement response information sent by the redetermined cluster head of the cluster to be merged; wherein the redetermined inter-cluster merge value of the cluster to be merged and the any cluster is second only to the previously determined inter-cluster merge value of the cluster to be merged and the any cluster.

[0056] After the cluster head of the cluster to be merged receives the cluster merge request information, if it does not agree to the cluster merge, it sends a cluster merge rejection response information to the node in any cluster, and the cluster merge rejection response information indicates that the cluster head of the cluster to be merged does not agree to the cluster merge. After the cluster head of any cluster receives the cluster merge rejection response information, it redefines the cluster to be merged in other clusters, and selects the cluster with the largest inter-cluster merging value from the other clusters except the cluster that sent the cluster merge rejection response information as the new cluster to be merged. If the cluster head of the newly determined cluster to be merged still does not agree to the merger, the above process is repeated until the cluster merge approval response information sent by the redetermined cluster head of the cluster to be merged is received.

[0057] The unmanned aerial vehicle network cluster maintenance method provided by the present invention has the following steps: after the cluster head of the sub-cluster to be merged receives the cluster merge request information, if it does not agree to the cluster merge, it sends a cluster merge rejection response information to the nodes in any sub-cluster; after the cluster head of any sub-cluster receives the cluster merge rejection response information, it redefines the sub-cluster to be merged in other sub-clusters until it receives the cluster merge approval response information sent by the cluster head of the redetermined sub-cluster to be merged, thereby improving the reliability of cluster merging.

[0058] Figure 3 This is one of the flow charts of cluster splitting in the UAV network cluster maintenance method provided by the present invention. Figure 3 As shown in the figure, the process of cluster splitting includes: Step 201: In response to the cluster head of any sub-cluster of the drone network learning that the number of nodes in the any sub-cluster exceeds a preset node number threshold, the cluster head of any sub-cluster determines a new cluster head for cluster splitting from the any sub-cluster.

[0059] Any cluster can be any cluster in the drone network. If the cluster head of any cluster in the drone network learns that the number of nodes in any cluster exceeds the preset node number threshold, the cluster head of any cluster determines a new cluster head for cluster splitting from any cluster. When the number of nodes in any cluster exceeds the preset node number threshold, it will be split into two clusters, and the new cluster head for cluster splitting will become the cluster head of the new cluster obtained after the split.

[0060] Step 202: The cluster head of any sub-cluster broadcasts cluster splitting prompt information to other nodes in any sub-cluster; wherein the cluster splitting prompt information includes information of a newly added cluster head of the cluster splitting.

[0061] The cluster head of any sub-cluster broadcasts cluster splitting prompt information to other nodes in any sub-cluster; wherein the cluster splitting prompt information includes information of a newly added cluster head for cluster splitting, which is used to indicate that the cluster will be split into two clusters.

[0062] Step 203: After receiving the cluster splitting prompt information, if the newly added cluster head agrees with the cluster splitting, it broadcasts the cluster splitting agreement information to other nodes in any sub-cluster.

[0063] After receiving the cluster splitting prompt information, if the newly added cluster head agrees to the cluster splitting, it broadcasts the cluster splitting consent information to other nodes in any sub-cluster except the newly added cluster head of the cluster splitting, to indicate that the newly added cluster head of the cluster splitting agrees to the cluster splitting and becomes the new cluster head.

[0064] Step 204: After receiving the agreed cluster splitting information, the nodes in any sub-cluster determine whether the cluster head is updated to the newly added cluster head of the cluster splitting by calculating the cluster splitting value. If it needs to be updated to the newly added cluster head of the cluster splitting, the cluster head information is updated.

[0065] Each node can only belong to one cluster and has one cluster head. After receiving the cluster split information, the node in any cluster determines whether the cluster head is updated to the newly added cluster head by calculating the cluster split value, where the cluster split value is used to help the members in the cluster decide their belonging after cluster splitting. If it needs to be updated to the newly added cluster head of cluster splitting, the cluster head is updated to the newly added cluster head of cluster splitting. If it does not need to be updated to the newly added cluster head of cluster splitting, it remains as the original cluster head.

[0066] The drone network cluster maintenance method provided by the present invention learns that the number of nodes in any sub-cluster exceeds a preset node number threshold by responding to the cluster head of any sub-cluster of the drone network, the cluster head of any sub-cluster determines a new cluster head for cluster splitting from any sub-cluster, and the cluster head of any sub-cluster broadcasts cluster splitting prompt information to other nodes in any sub-cluster; wherein the cluster splitting prompt information includes information of the new cluster head for cluster splitting, after the new cluster head for cluster splitting receives the cluster splitting prompt information, if it agrees to cluster splitting, it broadcasts the information of agreeing to cluster splitting to other nodes in any sub-cluster, after the nodes in any sub-cluster receive the cluster splitting information, it determines whether the cluster head is updated to the new cluster head for cluster splitting by calculating the cluster splitting value, if it needs to be updated to the new cluster head for cluster splitting, then updates the cluster head information, by splitting a larger cluster into two clusters, it is ensured that the throughput rate in the cluster will not be too low due to too many nodes and adapts to the topological changes driven by tasks in the drone disaster relief scenario, thereby further improving the performance of the drone hierarchical clustering network.

[0067] According to a drone network cluster maintenance method provided by the present invention, the cluster splitting value is expressed as:

[0068] in, Representation Node and nodes The cluster splitting value, Representation Node and nodes The distance Representation Node and nodes The speed difference, , Represents the weighting coefficient.

[0069] Representation Node and nodes The distance The smaller the value, the closer the two nodes are. Representation Node and nodes The speed difference, The smaller it is, the slower the relative speed between the two nodes is, and the longer the link connectivity can be maintained. , represents the weighting coefficient, + =1, can be set = =0.5.

[0070] The remaining nodes in the cluster will calculate the value, to determine the value to be added The cluster where the smaller cluster head is located. If the value is smaller, it will remain in the original cluster; if a node determines that the new cluster head If it is smaller, it will be added to the new cluster after the split.

[0071] The UAV network cluster maintenance method provided by the present invention improves the rationality of cluster splitting value determination by weighted calculation of cluster splitting values ​​according to the distance between nodes and the speed difference between nodes.

[0072] According to a drone network cluster maintenance method provided by the present invention, the cluster head of any sub-cluster determines a new cluster head for cluster splitting from any sub-cluster, including: the cluster head of any sub-cluster calculates the node weights of other nodes in any sub-cluster except itself, and selects the node with the smallest node weight as the new cluster head for cluster splitting.

[0073] In the process of cluster splitting, the original cluster head will find the node with the smallest node weight other than itself in the cluster as the cluster head of the new cluster after the split, that is, the newly added cluster head after the cluster split. The node weight is a parameter in the WCA algorithm. , expressed as:

[0074]

[0075] in, Representation Node The node weights of Representation Node The node difference of Representation Node The node degree, Indicates the number of nodes in the preset cluster. Representation Node The distance to neighbor nodes and Representation Node The relative moving speed with neighboring nodes and, Representation Node Power consumed, , , , Represents the weighting coefficient.

[0076] Figure 4 This is the second flow chart of cluster splitting in the UAV network cluster maintenance method provided by the present invention. Figure 4 The process of cluster splitting is given through specific examples. Figure 4 As shown, the method includes: Step 1: When the cluster If the size of the cluster exceeds the set split threshold (the preset node number threshold), the cluster head will start the cluster split process. will find all the other nodes in the cluster except itself. The node with the smallest (node ​​weight) , and select it as the cluster head of the second cluster formed after the split. Broadcast to other nodes in the cluster Message, indicating cluster is about to split, and It is the cluster head of the new cluster split out.

[0077] Step 2: Node receives After receiving the message, if the cluster agrees to split and become the new cluster head, Broadcast Message, to notify Agree to split and become the new cluster head, and also notify other nodes in the cluster.

[0078] Step 3: Other nodes in the cluster receive After the message, it will be calculated and Two cluster heads value (cluster splitting value), and added to The cluster where the smaller cluster head is located. If a node determines that the original cluster head of If the value is smaller, it will remain in the original cluster and no operation will be performed. of If it is smaller, it will join the second cluster formed after the split. To join the second cluster, send Request. Indicates that it will join the second cluster and notify It will leave the original cluster.

[0079] The UAV network cluster maintenance method provided by the present invention calculates the node weights of other nodes in any sub-cluster except itself through the cluster head of any sub-cluster, selects the node with the smallest node weight as the new cluster head for cluster splitting, thereby improving the rationality of determining the new cluster head for cluster splitting.

[0080] According to a method for maintaining a drone network cluster provided by the present invention, the method further includes: after the newly added cluster head of the cluster splitting receives the cluster splitting prompt information, if it does not agree with the cluster splitting, it broadcasts a cluster splitting rejection information to the nodes in any of the sub-clusters; after the cluster head of any of the sub-clusters receives the cluster splitting rejection information, it redetermines the newly added cluster head of the cluster splitting from any of the sub-clusters until it receives the cluster splitting agreement information sent by the redetermined newly added cluster head of the cluster splitting; wherein the node weight of the newly determined newly added cluster head of the cluster splitting is second only to the node weight of the previously determined newly added cluster head of the cluster splitting.

[0081] After receiving the cluster splitting prompt information, if the newly added cluster head of the cluster splitting does not agree with the cluster splitting, it broadcasts the cluster splitting rejection information to the nodes in any sub-cluster. The cluster splitting rejection information may include the information of the newly added cluster head of the cluster splitting and the indication information of rejecting the cluster splitting and becoming the new cluster head; after receiving the cluster splitting rejection information, the cluster head of any sub-cluster re-determines the newly added cluster head of the cluster splitting from any sub-cluster, and the node weight of the newly determined newly added cluster head of the cluster splitting is second only to the node weight of the previously determined newly added cluster head of the cluster splitting. If the newly determined newly added cluster head of the cluster splitting still does not agree with the cluster splitting, the above process is repeated until the newly determined newly added cluster head of the cluster splitting agrees with the cluster splitting information sent by the cluster splitting rejection information is received.

[0082] The drone network cluster maintenance method provided by the present invention broadcasts cluster splitting rejection information to nodes in any sub-cluster after a newly added cluster head receives cluster splitting prompt information. After the cluster head of any sub-cluster receives the cluster splitting rejection information, a new cluster head for cluster splitting is re-determined from any sub-cluster until the cluster splitting consent information is received from the re-determined new cluster head for cluster splitting, thereby improving the reliability of cluster splitting.

[0083] Figure 5 FIG. 1 is one of the flow diagrams of cluster head replacement in the UAV network cluster maintenance method provided by the present invention. Figure 5 As shown in Figure 1, the cluster head replacement process includes: Step 301: In response to the cluster head of any cluster of the drone network learning that there is a node in any cluster whose node weight is smaller than its own, the node with the smallest node weight is determined as the cluster head to be replaced.

[0084] Nodes with smaller weights are more suitable to be cluster heads. If the cluster head of any cluster of the UAV network learns that there is a node with a smaller weight than itself in any cluster, the node with the smallest weight is determined as the cluster head to be replaced.

[0085] Step 302: The cluster head of any cluster sends a cluster head replacement request to the cluster head to be replaced.

[0086] The cluster head of any sub-cluster sends a cluster head replacement request to the cluster head to be replaced. The cluster head replacement request is used to request to replace the cluster head of any sub-cluster with the cluster head to be replaced.

[0087] Step 303: After receiving the cluster head replacement request, if the cluster head to be replaced agrees to become the new cluster head, it broadcasts a message of agreeing to update the cluster head to other nodes in any sub-cluster.

[0088] After receiving the cluster head replacement request, if the cluster head to be replaced agrees to become the new cluster head, it broadcasts a cluster head update agreement message to other nodes in any sub-cluster. The cluster head update agreement message may include information of the cluster head to be replaced and indication information of agreeing to become the new cluster head.

[0089] Step 304: After receiving the message of agreeing to update the cluster head, the node in any sub-cluster updates the cluster head to the cluster head to be replaced.

[0090] After receiving the message of agreeing to update the cluster head, the node in any sub-cluster modifies the cluster head information and updates the cluster head to the cluster head to be replaced.

[0091] Figure 6 This is the second flow chart of cluster head replacement in the UAV network cluster maintenance method provided by the present invention. Figure 6 The process of cluster head replacement is given through a specific example. Figure 6 As shown in the figure, the process of cluster head replacement includes: Step 1: If the cluster head Detection of cluster Node The value is less than itself, indicating The node is more suitable to serve as the cluster head. The cluster head replacement process will be started. send Message, indicating cluster head The cluster head is about to be replaced and the The new cluster head.

[0092] Step 2: When Node receives After receiving the message, if the new cluster head is agreed to become the new cluster head, it will broadcast to the cluster Message, indicating Nodes agree to replace Become the new cluster leader.

[0093] Step 3: When other nodes receive After receiving the message, the cluster head is modified to node, The node no longer serves as the cluster head.

[0094] Through the above steps, the LP-WCA algorithm realizes the dynamic replacement process of cluster head nodes. The algorithm detects the node weights of member nodes in the cluster and selects nodes that are more suitable to serve as cluster heads, thereby improving the survival time of nodes and the life of the network. This mechanism allows cluster head nodes in the network to extend the life of the network by replacing new cluster heads after a certain amount of energy consumption.

[0095] The drone network cluster maintenance method provided by the present invention learns that there is a node with a node weight smaller than itself in any cluster in response to the cluster head of any cluster of the drone network, then determines the node with the smallest node weight as the cluster head to be replaced, and the cluster head of any cluster sends a cluster head replacement request to the cluster head to be replaced. After the cluster head to be replaced receives the cluster head replacement request, if it agrees to become the new cluster head, it broadcasts a message of consent to update the cluster head to other nodes in any cluster. After the nodes in any cluster receive the message of consent to update the cluster head, they update the cluster head to the cluster head to be replaced. By replacing inappropriate cluster head nodes, the problem of excessive energy consumption of cluster head nodes is solved, thereby improving the survival time of nodes and the life of the network, and further improving the performance of the drone hierarchical cluster network.

[0096] According to a drone network cluster maintenance method provided by the present invention, the method also includes: after the cluster head to be replaced receives the cluster head replacement request, if the cluster head to be replaced does not agree to become the new cluster head, it broadcasts a cluster head update refusal message to other nodes in any of the sub-clusters; after the cluster head of any of the sub-clusters receives the cluster head update refusal message, if it is learned that there are still nodes with node weights smaller than itself in any of the sub-clusters, it determines the node with the smallest node weight among the nodes that have not been selected as the cluster head to be replaced as the new cluster head to be replaced, until the cluster head agrees to update the cluster head message is received.

[0097] After receiving the cluster head replacement request, if the cluster head to be replaced does not agree to become the new cluster head, it broadcasts a cluster head update refusal message to other nodes in any sub-cluster. The cluster head update refusal message may include the information of the cluster head to be replaced and the indication information of refusing to become the new cluster head. After receiving the cluster head update refusal message, if the cluster head of any sub-cluster learns that there are still nodes with node weights less than its own in any sub-cluster, it determines the node with the smallest node weight among the nodes that have not been selected as the cluster head to be replaced as the new cluster head to be replaced. If the newly determined cluster head to be replaced still refuses to become the new cluster head, the above process is further repeated until the cluster head to be replaced receives the cluster head update consent message sent by the newly determined cluster head to be replaced.

[0098] After receiving the cluster head update refusal message, the cluster head of any sub-cluster can put the node that sent the cluster head update refusal message into the gray list. The nodes in the gray list will not be selected as the cluster head to be replaced in the cluster head replacement within a preset time, so as to avoid the waste of processing resources caused by the fact that after the cluster head is changed, it is still listed as the cluster head to be replaced due to its small node weight, but it is not suitable to be the cluster head due to the working state of the node and other reasons. After the preset time, the node can be deleted from the gray list.

[0099] The drone network cluster maintenance method provided by the present invention has the following steps: after the cluster head to be replaced receives a cluster head replacement request, if the cluster head to be replaced does not agree to become the new cluster head, it broadcasts a cluster head update refusal message to other nodes in any sub-cluster; after the cluster head of any sub-cluster receives the cluster head update refusal message, if it is known that there are still nodes with node weights smaller than itself in any sub-cluster, it determines the node with the smallest node weight among the nodes that have not been selected as the cluster head to be replaced as the new cluster head to be replaced, thereby improving the reliability of the cluster head replacement process.

[0100] Figure 7 This is a comparison chart of the average residual energy consumption of all network nodes in the dual-channel system between the LP-WCA algorithm and the WCA algorithm. Figure 7 As shown in the figure, it can be seen that the node energy consumption in the LP-WCA algorithm is slower than that in the WCA algorithm. In addition, the energy consumption rate of the two algorithms remains unchanged in the early stage of the network, but in the later stage of the network, the node energy consumption rate suddenly slows down until the network dies. This is because in the early stage of network operation, the number of neighbor nodes of the node is constant, and the amount of data sent and received by each node remains unchanged, so the energy is consumed evenly; but when most of the nodes in the network die, the number of neighbors of the node decreases, resulting in the node receiving less data from neighbor nodes, so the energy consumption rate slows down significantly.

[0101] Figure 8 Figure 1 is a schematic diagram of the simulation results of the number of clusters changing over time in the dual-channel system when the LP-WCA algorithm and the WCA algorithm are running in the network. Figure 8As shown in the figure, we can see that from the beginning of network simulation to the time when the number of clusters begins to decrease, the number of clusters of the LP-WCA algorithm is about 30% less than that of the WCA algorithm. The simulation proves that the LP-WCA algorithm can effectively reduce the number of clusters in network operation during the cluster maintenance phase. In addition, when the LP-WCA algorithm is running on a network with 100 nodes, the number of clusters 8 ≤ M ≤ 15 meets the index requirements.

[0102] Fig. 9 The figure is a comparison of the network task cycles of the LP-WCA algorithm and the WCA algorithm in a dual-channel system. The task cycles of the drone swarm network are divided into five categories: (1) The network optimal period is the time from the start of the network to the first death of the node; (2) The network efficient period is the time from the first death of the node to the death of 10% of the nodes; (3) The network effective period is the time from the death of 10% of the nodes to the death of 50% of the nodes; (4) The network available period is the time from the death of 50% of the nodes to the death of 90% of the nodes; (5) The network unstable period is the time from the death of 90% of the nodes to the death of all the nodes. Fig. 9 It can be seen that the network optimal period of the LP-WCA algorithm is improved by about 180% compared with the WCA algorithm, the network high efficiency period of the LP-WCA algorithm is improved by about 100% compared with the WCA algorithm, the network effective period is slightly lower than the WCA algorithm by 30%, the network availability period is improved by about 25% compared with the WCA algorithm, and the network unstable period is reduced by about 50% compared with the WCA algorithm. From the above analysis, it can be seen that the LP-WCA algorithm improves the optimal period and high efficiency period of the network in the dual-channel system, reduces the network unstable period, and improves the performance of the drone swarm network in executing tasks.

[0103] The present invention designs a new clustering algorithm, the LP-WCA algorithm (Low Power Weighted Cluster Algorithm), which optimizes the cluster merging, splitting, and cluster head replacement processes in the cluster maintenance phase. First, the LP-WCA algorithm designs a cluster merging algorithm to address the problem of fewer nodes in the WCA algorithm cluster, solving the problem of short network survival time in the cluster maintenance phase; then, in order to ensure that the service channel in the cluster does not have too low throughput due to too many nodes and to adapt to task-driven topology changes in disaster relief scenarios, the LP-WCA algorithm designs an efficient cluster splitting algorithm; finally, the LP-WCA algorithm designs an efficient cluster head replacement algorithm to address the problem of too fast energy consumption of cluster head nodes in the classic WCA algorithm in a dual-channel system, to solve the problem of too fast death of initial nodes.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for maintaining a drone network cluster, characterized in that: include: In response to a cluster head of any sub-cluster of the drone network learning that there are other sub-clusters around the cluster whose total number of nodes after merging with the any sub-cluster does not exceed the number of nodes in the preset cluster, the cluster head of any sub-cluster determines a cluster to be merged from the other sub-clusters; The cluster head of any sub-cluster sends cluster merging request information to the cluster head of the sub-cluster to be merged; After receiving the cluster merging request information, if the cluster head of the to-be-merged sub-cluster agrees to the cluster merging, the cluster head sends a cluster merging agreeing response information to the nodes in any of the sub-clusters; After receiving the cluster merging agreement response information, the node in any sub-cluster changes the cluster head to the cluster head of the sub-cluster to be merged.

2. The method for maintaining a drone network cluster according to claim 1, characterized in that: The cluster head of any cluster determines the cluster to be merged from the other clusters, including: the cluster head of any cluster calculates the inter-cluster merging value with the other clusters, and selects the cluster with the smallest inter-cluster merging value as the cluster to be merged; the inter-cluster merging value is expressed as: ; in, Represents clustering and clustering The inter-cluster merge value between, Represents clustering and clustering The number of nodes with communication links in represents the number of nodes in the preset cluster, Represents clustering The number of nodes, Represents clustering The number of nodes, Represents clustering Cluster Head and clustering The distance between cluster heads, , , Represents the weighting coefficient.

3. The method for maintaining a drone network cluster according to claim 2, characterized in that: The method further comprises: After receiving the cluster merging request information, if the cluster head of the sub-cluster to be merged disagrees with the cluster merging, it sends a cluster merging rejection response information to the node in any sub-cluster; After receiving the cluster merging rejection response information, the cluster head of any cluster redetermines the cluster to be merged in the other clusters until receiving the cluster merging approval response information sent by the redetermined cluster head of the cluster to be merged; The re-determined inter-cluster merging value between the to-be-merged cluster and the any cluster is second only to the previously determined inter-cluster merging value between the to-be-merged cluster and the any cluster.

4. The method for maintaining a drone network cluster according to claim 1, characterized in that: The method further comprises: In response to a cluster head of any sub-cluster of the drone network learning that the number of nodes in any sub-cluster exceeds a preset node number threshold, the cluster head of any sub-cluster determines a new cluster head for cluster splitting from the any sub-cluster; The cluster head of any sub-cluster broadcasts cluster splitting prompt information to other nodes in any sub-cluster; wherein the cluster splitting prompt information includes information of a newly added cluster head of the cluster splitting; After receiving the cluster splitting prompt information, if the newly added cluster head agrees with the cluster splitting, the newly added cluster head broadcasts the cluster splitting agreement information to other nodes in any sub-cluster; After receiving the agreed cluster splitting information, the nodes in any sub-cluster determine whether the cluster head is updated to the newly added cluster head of the cluster splitting by calculating the cluster splitting value. If it needs to be updated to the newly added cluster head of the cluster splitting, the cluster head information is updated.

5. The method for maintaining a drone network cluster according to claim 4, characterized in that: The cluster split value is expressed as: ; in, Representation Node and nodes The cluster splitting value, Representation Node and nodes The distance Representation Node and nodes The speed difference, , Represents the weighting coefficient.

6. The method for maintaining a drone network cluster according to claim 4, characterized in that: The cluster head of any sub-cluster determines a newly added cluster head for cluster splitting from any sub-cluster, including: The cluster head of any sub-cluster calculates the node weights of other nodes in the sub-cluster except itself, and selects the node with the smallest node weight as the newly added cluster head for cluster splitting.

7. The method for maintaining a drone network cluster according to claim 6, characterized in that: The node weight is expressed as: ; ; in, Representation Node The node weights of Representation Node The node difference of Representation Node The node degree, represents the number of nodes in the preset cluster, Representation Node The distance to neighbor nodes and Representation Node The relative moving speed with neighboring nodes and, Representation Node Power consumed, , , , Represents the weighting coefficient.

8. The method for maintaining a drone network cluster according to claim 6, characterized in that: The method further comprises: After receiving the cluster splitting prompt information, if the newly added cluster head disagrees with the cluster splitting, it broadcasts cluster splitting rejection information to the nodes in any sub-cluster; After receiving the cluster splitting rejection information, the cluster head of any sub-cluster redetermines the cluster splitting newly added cluster head from any sub-cluster until receiving the cluster splitting approval information sent by the redetermined cluster splitting newly added cluster head; The node weight of the newly determined cluster head added after cluster splitting is second only to the node weight of the newly determined cluster head added after cluster splitting.

9. The method for maintaining a drone network cluster according to claim 6, characterized in that: The method further comprises: In response to a cluster head of any sub-cluster of the drone network learning that there is a node in any sub-cluster whose node weight is smaller than its own, determining the node with the smallest node weight as the cluster head to be replaced; The cluster head of any sub-cluster sends a cluster head replacement request to the cluster head to be replaced; After receiving the cluster head replacement request, if the cluster head to be replaced agrees to become the new cluster head, the cluster head to be replaced broadcasts a message of agreeing to update the cluster head to other nodes in any of the sub-clusters; After receiving the cluster head update agreement message, the nodes in any sub-cluster update the cluster head to be replaced.

10. The method for maintaining a drone network cluster according to claim 9, characterized in that: The method further comprises: After receiving the cluster head replacement request, if the cluster head to be replaced does not agree to become the new cluster head, it broadcasts a cluster head update refusal message to other nodes in any of the sub-clusters; After the cluster head of any of the clusters receives the message of refusing to update the cluster head, if it is learned that there are still nodes with node weights smaller than its own in any of the clusters, it determines the node with the smallest node weight among the nodes that have not been selected as the cluster head to be replaced as the new cluster head to be replaced, until it receives the message of agreeing to update the cluster head.