Pre-planning cluster head pushing method for rapid establishment of directional network of unmanned aerial vehicle group

Abstract

The invention discloses a pre-planning cluster head pushing method for rapid establishment of a directional network of an unmanned aerial vehicle group, and the method comprises the steps: obtaining a potential candidate cluster first choice set through screening according to the comprehensive energy value of each unmanned aerial vehicle node, and obtaining a candidate cluster first choice set according to the single-beam coverage capability of a directional antenna; redundant unmanned aerial vehicle nodes covering the overlapping area in the potential cluster head candidate set are eliminated to obtain candidate cluster heads, then quantitative evaluation of network communication quality, energy state and topological stability is carried out on the candidate cluster heads, and the unmanned aerial vehicle node with the highest quantitative evaluation score is used as the cluster head; and the unmanned aerial vehicle node with the second highest quantitative evaluation score is used as a standby cluster head, so that the cluster head is determined timely and accurately, and the beam resource utilization rate is improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) swarm ad hoc network communication technology, and in particular to a pre-planned cluster leader election method and apparatus for rapid network construction of UAV swarm directional networks. Background Technology

[0002] In the field of UAV swarm ad hoc network communication, the cluster head election mechanism is a core technology that determines network performance. Current methods typically employ pre-launch customized cluster heads or random election mechanisms. However, in high-speed, narrow-beam phased array networking scenarios, methods that determine cluster heads solely based on node location or energy indicators lack multi-dimensional collaborative optimization. Furthermore, in high-speed flight scenarios, node mobility leads to a surge in the probability of cluster head failure. Therefore, how to determine cluster heads in a timely and accurate manner to improve beam resource utilization has become an urgent problem to be solved. Summary of the Invention

[0003] This invention provides a pre-planned cluster head selection method and apparatus for rapid network construction of directional networks for UAV swarms, in order to solve the problem that existing methods cannot determine cluster heads in a timely and accurate manner, resulting in wasted beam resources and coverage blind spots.

[0004] In a first aspect, the present invention provides a pre-planned cluster leader election method for rapid network construction of directional networks for UAV swarms. The method includes: constructing a directional network topology of the UAV swarm based on the initial position coordinates, movement speed and direction parameters of each UAV node in the UAV swarm. Based on the directional network topology, each UAV node calculates its own comprehensive energy value based on information from its neighboring UAV nodes. Unmanned aerial vehicle (UAV) nodes with a comprehensive energy value greater than a preset energy threshold are selected as potential candidate cluster heads. Based on the single-beam coverage capability of the directional antenna, redundant UAV nodes covering overlapping areas within the potential candidate cluster heads are eliminated to obtain the candidate cluster heads; The candidate cluster heads are quantitatively evaluated for energy status, network communication quality, and topology stability. The UAV node with the highest quantitative evaluation score is selected as the cluster head, and the UAV node with the second highest quantitative evaluation score is selected as the backup cluster head. During the flight of the drone swarm, the network topology status is monitored in real time, and the cluster head is upgraded when the cluster head replacement trigger condition is met. The neighboring drone nodes of each drone node are determined based on a preset minimum distance principle. The distance variance is used to characterize the centrality of the cluster heads. For speed consistency, used to characterize drone nodes The speed difference with its neighboring drone nodes The horizontal angle is used to characterize the delay in establishing a communication connection between UAV nodes during directional antenna polling. The weights used to evaluate the overall performance of drone nodes in terms of location, The weights used to evaluate the overall performance of drone nodes in terms of speed consistency The weights are used to evaluate the overall performance of UAV nodes in terms of distribution azimuth.

[0005] Optionally, the distance variance is ,in, Indicates drone node and drone nodes The Euclidean distance between them; Indicates drone node The average distance to the other drone nodes in the cluster; This represents the set of all drone nodes within the cluster; The speed consistency is Including directional differences and speed difference ,in, Indicates drone node The velocity vector; This represents the average velocity vector of the neighboring drone nodes. For directional differences and This is a constant representing the velocity difference. The horizontal included angle is Where threshold is the horizontal angle threshold. For drone nodes With drone nodes The horizontal angle between them, and , Indicates drone node and drone nodes The difference in horizontal coordinates; Indicates drone node and drone nodes The vertical coordinate difference.

[0006] Optionally, based on the single-beam coverage capability of the directional antenna, redundant UAV nodes covering overlapping areas within the potential candidate cluster heads are excluded to obtain candidate cluster heads, including: Based on the single-beam coverage capability of the directional antenna, the beam coverage range is calculated. Then, based on a preset beam avoidance strategy, the beam pointing is controlled by adjusting the phase of the phased array antenna to prevent multiple UAV nodes from simultaneously entering the network under the same beam. Finally, the potential candidate cluster heads are screened to obtain the candidate cluster heads.

[0007] Optionally, the candidate cluster heads are subjected to a quantitative evaluation of their energy state, network communication quality, and topology stability, including: according to For each candidate cluster head, a quantitative evaluation of its energy state, network communication quality, and topology stability is performed. Remaining battery power Weighted by remaining battery power. For network communication quality, As a weight for network communication quality, For topological stability, These are the topological stability weights.

[0008] Optionally, the real-time monitoring of network topology status, triggering cluster head upgrades when cluster head replacement triggering conditions are met, includes: based on The system calculates the current network topology and transmission energy consumption of the drone swarm in real time, and determines whether the cluster head needs to be upgraded based on the calculation results. If so, it triggers the cluster head upgrade. in, Let be the number of neighboring drone nodes of drone node i. This represents the average connectivity of drone nodes. For the energy consumption of transmission between drone nodes, The remaining total energy of the drone node, For network connectivity weights, Transmission energy consumption weight.

[0009] Optionally, the real-time monitoring of network topology status, triggering cluster head upgrades when cluster head replacement triggering conditions are met, includes: Determine the probability of a cluster leaving the cluster by examining the cluster head. ,in, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of drone nodes that have already established a blockchain. For the current drone node The ratio of the damaged value to the normal value of the link. The indicator for adjusting the quality of the task; if If the value is less than the critical value Q and the current cluster head is not faulty, then the weight values ​​of each backup cluster head are calculated based on the current cluster head to become the nominated cluster head. And determine the next cluster head; if If the value is less than the critical value Q and the current cluster is faulty, then each backup cluster head calculates its own weight value to become the nominated cluster head. And through weight values The order determines the next cluster head.

[0010] Optionally, the weight value for each alternate cluster head to become the nominated cluster head can be calculated based on the current cluster head. And determine the next cluster head, including: Current cluster head based on Calculate the weight value for each backup cluster head to become the nominated cluster head. ,in, The total number of members in the cluster. Weights for the number of drone nodes in the established blockchain. Assign weights to the drone node addresses. This refers to the drone node address number; The calculated weight values Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head, and all alternate cluster heads are informed accordingly.

[0011] Optionally, each of the backup cluster heads calculates its own weight value for becoming the nominated cluster head. And through weight values The sorting process determines the next cluster head, including: Each backup cluster head is based on Calculate the weight value for each to become the head of the nominated cluster. And by weighting all backup cluster heads. Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head; in, For the remaining energy weight, It is the ratio of energy consumed to total energy. For the current drone node The ratio of the damaged value to the normal value of the link. Distance weight between the standby cluster head and the original failed cluster head. This represents the distance between the backup cluster head and the original faulty cluster head.

[0012] Secondly, the present invention provides a pre-planned cluster leader election device for rapid network construction of UAV swarm-oriented networks, the device comprising: The first processing unit is used to construct the directional network topology of the UAV cluster based on the initial position coordinates, movement speed and direction parameters of each UAV node in the UAV cluster. The second processing unit is used to calculate the comprehensive energy value of each UAV node based on its neighboring UAV node information according to the directional network topology. Drone nodes with a comprehensive energy value greater than a preset energy threshold are selected as potential candidate cluster heads. The neighboring drone nodes of each drone node are determined based on a preset minimum distance principle. The distance variance is used to characterize the centrality of the cluster heads. For speed consistency, used to characterize drone nodes The speed difference with its neighboring drone nodes The horizontal angle is used to characterize the delay in establishing a communication connection between UAV nodes during directional antenna polling. Used to evaluate the overall performance of drone nodes in terms of location. Used to evaluate the overall performance of drone nodes in terms of speed consistency. Used to evaluate the overall performance of UAV nodes in terms of distribution azimuth angle; The third processing unit is used to eliminate redundant UAV nodes covering overlapping areas within the potential candidate cluster heads based on the single-beam coverage capability of the directional antenna, thereby obtaining candidate cluster heads. The quantitative evaluation unit is used to quantitatively evaluate the energy state, network communication quality and topology stability of the candidate cluster heads, and selects the UAV node with the highest quantitative evaluation score as the cluster head and the UAV node with the second highest quantitative evaluation score as the backup cluster head. The monitoring unit is used to monitor the network topology status in real time during the flight of the drone swarm, and to trigger the cluster head upgrade when the cluster head replacement trigger condition is met.

[0013] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the pre-planned cluster leader election method for rapid network construction of UAV swarm-oriented networks as described above.

[0014] The beneficial effects of this invention are as follows: This invention selects potential candidate cluster heads based on the comprehensive energy value of each UAV node, and eliminates redundant UAV nodes in overlapping coverage areas within the potential candidate cluster heads based on the single-beam coverage capability of the directional antenna. Then, by quantitatively evaluating the network communication quality, energy status, and topology stability of the candidate cluster heads, the UAV node with the highest quantitative evaluation score is selected as the cluster head, and the UAV node with the second highest quantitative evaluation score is selected as the backup cluster head. This allows for timely and accurate determination of cluster heads, thereby improving beam resource utilization.

[0015] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0016] 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 invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating a cluster head election method for rapid network construction in a directional network for unmanned aerial vehicle (UAV) swarms, provided by an embodiment of the present invention. Figure 2 This is a flowchart of the cluster head election method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the node changes in cluster head election provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the cluster head's self-determination and decoupling process for initiating network reconstruction, provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of the process by which the backup cluster head determines its own departure from initiating network reconstruction, provided in an embodiment of the present invention. Figure 6 This is a schematic diagram of a cluster head election device for rapid network construction of directional networks for UAV swarms, provided in an embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of the invention.

[0018] To address the issues of the disconnect between beam spatial characteristics and resource allocation in existing UAV swarm ad hoc networks, the lack of multi-dimensional collaborative optimization in existing cluster head selection which is based solely on node location or energy indicators, the increased probability of cluster head failure due to node mobility in high-speed flight scenarios, and the difficulty in meeting real-time requirements due to reliance on global information synchronization in existing reselection mechanisms.

[0019] To address the aforementioned problems, embodiments of the present invention provide a pre-planned cluster leader election method for rapid network construction in UAV swarm-oriented networks. See [link to relevant documentation]. Figure 1 The method includes: S101. Based on the initial position coordinates, movement speed and direction parameters of each UAV node in the UAV cluster, the directional network topology of the UAV cluster is constructed. Specifically, in this embodiment of the invention, the topology of the drone swarm is first constructed, and then the cluster head is selected based on the topology. Of course, this topology is updated in real time as the drone nodes move, so the cluster head also needs to be updated in real time.

[0020] In specific implementation, embodiment S101 of the present invention specifically includes: After the drone swarm is started, all drone nodes (hereinafter referred to as nodes) are triggered to broadcast beacon frames through a preset communication protocol. Then, each node collects its own three-dimensional position (including longitude, latitude, and degrees), velocity vector (including horizontal and vertical velocity), flight direction (heading angle), and other information. It also periodically exchanges data through the TDMA time division multiplexing mechanism to ensure that the information of neighboring nodes can be synchronized in real time. Based on signal strength and distance estimation, a dynamic adjacency list is built through the neighbor discovery mechanism to record the list of neighboring nodes and their real-time status. Finally, the topology of the drone swarm is established and updated in real time according to the changes of the nodes.

[0021] S102. Based on the directional network topology, each UAV node calculates its own comprehensive energy value based on the information of its neighboring UAV nodes. Unmanned aerial vehicle (UAV) nodes with a comprehensive energy value greater than a preset energy threshold are selected as potential candidate cluster heads. Since there are usually multiple potential candidate cluster heads, it can also be called the preferred set of potential candidate clusters; Specifically, in this embodiment of the invention, the comprehensive energy value of a node is calculated based on neighboring drone nodes, and drone nodes with comprehensive energy values ​​greater than a preset energy threshold are selected as potential candidate cluster heads. Then, candidate cluster heads are obtained by further filtering from these potential candidate cluster heads.

[0022] In specific implementation, the distance variance of the embodiments of the present invention Used to characterize the centrality of the cluster head. Indicates drone node and drone nodes The Euclidean distance between them; Indicates drone node The average distance to the other drone nodes in the cluster; This represents the set of all drone nodes within the cluster; Speed ​​consistency of embodiments of the present invention Used for quantifying drone nodes Differences in speed with neighbors, including differences in direction and speed difference ,in, Represents a node The velocity vector; This represents the average velocity vector of neighboring nodes; Indicates consistency in direction; Indicates the difference in speed; Horizontal angle of the present invention This is used to characterize the delay in establishing a communication connection during directional antenna polling, where threshold is the horizontal angle threshold. , Represents a node and nodes The difference in horizontal coordinates; Represents a node and nodes The vertical coordinate difference; For nodes and nodes The horizontal angle.

[0023] in, The weights used to evaluate the overall performance of drone nodes in terms of location, The weights used to evaluate the overall performance of drone nodes in terms of speed consistency The weights are used to evaluate the overall performance of UAV nodes in terms of distribution azimuth.

[0024] It should be noted that, in the embodiments of the present invention, a preset energy threshold and , and The values ​​are determined through experiments. Of course, in specific implementation, those skilled in the art can set them arbitrarily according to actual needs, but this invention will not discuss this in detail.

[0025] In specific implementation, the centrality evaluation in this embodiment of the invention is as follows: Smaller nodes are closer to the center of the network topology and have a stronger chance of being cluster head candidates. In specific implementation, the embodiments of the present invention determine neighboring nodes based on a preset minimum distance principle. That is, the distances between each node and its nearest connected node are added together, and the node with the smallest sum is taken as a cluster. The minimum distance, i.e. the sum of the distances between nodes, can be specifically set according to the number of nodes in the drone swarm, the number of clusters, and the specific control timeliness, etc. For example, the minimum distance can be set to 20 meters, i.e., the sum of the distances between all nodes in the cluster is 20 meters. Nodes exceeding this value will be assigned to other clusters. In specific implementation, the method for screening cluster head candidate sets according to the present invention includes: preliminary screening: selecting from neighboring nodes. The smallest 10% of nodes are selected as the candidate set; the candidate node set N = {1, 2, 3, ..., N} represents the position of node i. , ; Beamforming and coverage: ,in, The main lobe pointing angle of the beam. For the signal wavelength, Array element spacing; The phase offset of the nth element is .

[0026] Collision detection: If other candidate nodes exist within the coverage area of ​​the current beam of a candidate node, then through... To eliminate possibilities.

[0027] S103. Based on the single-beam coverage capability of the directional antenna, eliminate redundant UAV nodes covering overlapping areas within the potential candidate cluster heads to obtain the candidate cluster heads; Specifically, in this embodiment of the invention, the beam coverage range is calculated based on the single-beam coverage capability of the directional antenna, and the beam pointing is controlled by adjusting the phase of the phased array antenna based on a preset beam avoidance strategy, so as to avoid multiple UAV nodes entering the network at the same time under the same beam, and the potential candidate cluster heads are screened to determine the final candidate cluster heads.

[0028] The beam avoidance strategy achieves dynamic beam pointing by adjusting the phase of the phased array antenna. , Where is the signal wavelength, N is the number of array elements, and d is the element spacing.

[0029] S104. Quantitatively evaluate the energy state, network communication quality, and topology stability of the candidate cluster heads, and select the UAV node with the highest quantitative evaluation score as the cluster head, and select the UAV node with the second highest quantitative evaluation score as the backup cluster head. according to For each candidate cluster head, a quantitative evaluation of its energy state, network communication quality, and topology stability is performed. The remaining charge of the candidate cluster head. Weighted by remaining battery power. For the network communication quality of candidate cluster heads, As a weight for network communication quality, For the topological stability of candidate cluster heads, These are the topological stability weights. In the embodiments of this invention, each weight can be determined experimentally, and those skilled in the art can set them specifically; this invention does not impose any specific limitations on this.

[0030] It should be noted that in the embodiments of the present invention, there may be multiple drone nodes with the same quantitative evaluation score. When there are multiple highest quantitative evaluation scores, the present invention can randomly select a drone node with the highest quantitative evaluation score as the cluster head. In more common cases, there are multiple drone nodes with the second highest quantitative evaluation scores, or multiple drone nodes may have very close quantitative evaluation scores, all within a common score range, such as 8.1, 8.5, 8.7, 8.9, etc. They are all within an acceptable scoring range [8-9] (the specific scoring range can be arbitrarily set according to actual needs; the present invention is merely illustrating this with an example). In this case, multiple backup cluster heads can be set. By setting multiple backup cluster heads, an optimal cluster head can be selected from multiple cluster heads during subsequent cluster head upgrades.

[0031] S105. During the flight of the UAV swarm, monitor the network topology status in real time, and trigger the cluster head upgrade when the cluster head replacement trigger condition is met.

[0032] In specific implementation, embodiments of the present invention calculate the current network topology and transmission energy consumption within the cluster based on the ground control terminal or the current cluster head during cluster flight. And determine whether the cluster head needs to be upgraded, among which, Let be the number of neighbors of drone node i. For network connectivity weights, For node transmission energy consumption, for, This represents the average connectivity of the nodes. This represents the total remaining energy of the node. If the calculated... The value is lower than the preset value If the threshold is reached, a cluster head upgrade will be triggered.

[0033] In addition, in specific implementation, those skilled in the art can also set various cluster head upgrade triggering conditions according to actual needs. For example, during the flight of a drone swarm, the probability of the cluster head leaving the swarm can be determined by the current cluster head. ,in, To decouple from the cluster weight value, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. The indicator for adjusting the quality of the task; If calculation Less than the critical value If the current cluster head is not faulty, then the weight value of the nominated cluster head for each backup cluster head is calculated based on the current cluster head. And determine the next cluster head; if Less than the critical value If the current cluster fails, the weight values ​​for each cluster to become the nominated cluster leader are calculated using the backup cluster leaders. And determine the next cluster head by sorting; In other words, for drone swarm flight, the first step is through calculation. This involves quantitatively evaluating the energy state, network communication quality, and topology stability of each candidate cluster head to determine the cluster head and backup cluster heads, and then... The system calculates the current network topology and transmission energy consumption of the drone swarm in real time, determines whether the cluster head needs to be upgraded based on the calculation results, and assesses the probability of the cluster head leaving the swarm. This determines whether a cluster head upgrade is needed. When any condition for a cluster head upgrade is met, the upgrade is triggered, thus identifying the cluster head in a timely and accurate manner, thereby improving beam resource utilization.

[0034] Specifically, in this embodiment of the invention, cluster head upgrade can be determined based on different cluster head upgrade conditions, such as the current network topology state within the cluster and transmission energy consumption calculations. ,if If the value exceeds a preset calculation threshold, a cluster head upgrade is triggered, or the probability of the cluster head leaving the cluster is determined by the cluster head itself. ,if If the value is less than the critical value Q, the cluster head upgrade is triggered. Of course, in specific implementation, those skilled in the art can also set other upgrade rules according to actual needs to ensure the accuracy of the cluster head, thereby achieving efficient self-control of the drone swarm and ultimately completing the predetermined mission objectives through the drone swarm.

[0035] In this embodiment of the invention, the weight value of the proposed cluster head for each backup cluster head is calculated based on the current cluster head, and the next cluster head is determined, including: Current cluster head based on Calculate the weight value for each backup cluster head to become the nominated cluster head. ,in, The total number of members in the cluster. Weights for the number of drone nodes in the established blockchain. Assign weights to the drone node addresses. This refers to the drone node address number; The calculated weight values Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head, and the weight value is set accordingly. The second-highest spare cluster head becomes the next spare cluster head, and all spare cluster heads are notified at the same time.

[0036] It should be noted that, in specific implementation, this invention uses weight values... The next few highest-ranking backup cluster heads are used as the next backup cluster heads, so that the cluster heads for the next subsequent generation can be better determined. In simple terms, this embodiment of the invention uses... The highest-ranking alternate cluster head will become the next cluster head, and will The maximum number of backup cluster heads (e.g., 3) are used as backup cluster heads for the next term, so that the next cluster head can be better determined from these backup cluster heads.

[0037] Specifically, in this embodiment of the invention, the weight value for each backup cluster head to become the nominated cluster head is calculated one by one by the cluster heads. According to priority and Order of magnitude difference After each calculation is completed, the results are sorted and communicated to each backup cluster head. Upon receiving the message, the backup cluster head determines whether it is the nominated cluster head. If so, it initiates a handshake with each member in the network management time slot of the original cluster head after N seconds. If it is not the new cluster head, it waits for the new cluster head to handshake with it. The new cluster network is established after each member responds to the handshake.

[0038] Furthermore, the embodiments of the present invention describe calculating the weight values ​​of each backup cluster head to become a nominated cluster head. And determine the next cluster head by sorting, including: each backup cluster head based on Calculate the weight value for each to become the head of the nominated cluster. And by weighting all backup cluster heads. Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head, and the weight value is set accordingly. The second-highest spare cluster head becomes the next spare cluster head; It should be noted that, in specific implementation, this invention uses weight values... The next few backup cluster heads are used as the next backup cluster heads, so that the cluster heads for the next subsequent iteration can be better determined. For example, the next few backup cluster heads can be... The highest-ranking alternate cluster head will become the next cluster head, and will The maximum number of backup cluster heads (e.g., 3) are used as backup cluster heads for the next term, so that the next cluster head can be better determined from these backup cluster heads.

[0039] Specifically, in this embodiment of the invention, each backup cluster head calculates its weight value for becoming a nominated cluster head based on a comprehensive metric function for autonomously nominated cluster heads. This data is then sent to other backup cluster heads, and the weight values ​​of all nodes are obtained through statistical analysis. And sorted, the autonomously recommended cluster head comprehensive metric function is as follows: ,in, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. Distance weight between the standby cluster head and the original failed cluster head. The standby cluster head determines its own weight value based on the distance between the standby cluster head and the original failed cluster head. After ranking first, it updates its role to cluster head. N seconds later, in the network management time slot of the original cluster head, it initiates a handshake with each member. After each member responds to the handshake, the new cluster network is established.

[0040] In general, the method described in the embodiments of the present invention can realize the pre-planning and dynamic recommendation of cluster heads and backup cluster heads in directional networks. By accurately recommending cluster heads, the network can be built quickly, thereby improving the utilization rate of beam resources.

[0041] The following will combine Figures 2-5 The method described in the embodiments of the present invention will be explained and illustrated in detail through a specific example: Existing dynamic clustering technologies are based on node mobility and reduce the impact of topology changes by updating cluster heads in real time. However, they do not consider beam space characteristics, resulting in limited resource utilization. Furthermore, while multi-dimensional resource allocation exists in a joint spatial-temporal resource allocation model, it focuses on downlink optimization and fails to address beam collisions caused by multi-node competition in the uplink. This invention provides a pre-planned cluster head election method for rapid network construction in UAV swarm directional networks. The method includes: See Figure 2 The first step in this embodiment of the invention is to initialize the nodes. After the drone swarm is started, it exchanges data such as position, speed and direction information through data frames. System startup: After the drone swarm starts, all nodes broadcast beacon frames through a preset communication protocol; Information acquisition: Each node acquires information such as its own three-dimensional position (longitude, latitude, altitude), velocity vector (horizontal velocity, vertical velocity), and flight direction (heading angle); Data synchronization: Nodes periodically exchange data through the TDMA time-division multiplexing mechanism to ensure real-time synchronization of information between neighboring nodes; Neighbor discovery: Based on signal strength and distance estimation, a dynamic adjacency list is constructed to record the list of neighboring nodes and their real-time status.

[0042] Secondly, a comprehensive metric calculation is performed, with each node calculating its own comprehensive metric based on its neighbor information. ; In the embodiments of the present invention, the distance variance sub-index The formula used to characterize the centrality of the cluster head is as follows: ,in, Represents a node and nodes The Euclidean distance between them; Represents a node The average distance to the other nodes in the cluster; This represents the set of all nodes within a cluster.

[0043] The smaller the value, the more likely it is to be a node. When acting as a cluster head, the more evenly the distribution of the remaining nodes, the more balanced the communication latency.

[0044] Speed ​​consistency score in the embodiments of the present invention Used for quantization nodes Differences in speed with neighboring nodes, including differences in direction and magnitude: , ,in, Represents a node The velocity vector; This represents the average velocity vector of neighboring nodes; Indicates directional consistency; the closer to 1, the more consistent the directions. This indicates the difference in speed magnitude. Ultimately, a speed consistency score is calculated. The calculation is as follows: .

[0045] Horizontal angle in the embodiments of the present invention The calculation steps for characterizing the delay in establishing a communication connection during directional antenna polling include: ,in, Represents a node and nodes The difference in horizontal coordinates; Represents a node and nodes The vertical coordinate difference; For nodes and nodes The horizontal angle.

[0046] In this embodiment of the invention, the horizontal angles of all nodes are calculated, and horizontal angles greater than a threshold are negativeed: This formula indicates that the smaller the horizontal angle value, the higher the distribution aggregation of the other nodes when the node acts as the cluster head.

[0047] Centrality evaluation in this invention embodiment: Smaller nodes are closer to the center of the network topology and have a stronger chance of becoming cluster head candidates.

[0048] Then, cluster head candidate screening is performed, and the node with the smallest distance that satisfies the phased array beam pointing constraint is selected as the cluster head candidate. Preliminary screening: Selecting from neighboring nodes The smallest 10% of nodes are used as the candidate set; Candidate node set N = {1, 2, 3, ..., N}: Position of node i

[0049] ; Beamforming and coverage: ; in, The main lobe pointing angle of the beam. For the signal wavelength, Array element spacing, Phase offset of the nth element .

[0050] Collision detection: If other candidate nodes exist within the coverage area of ​​the current beam of a candidate node, they are excluded according to the following rules: ; Next, pre-planned cluster heads and backup cluster heads are performed. The network communication quality, energy state, and topology stability of the candidate cluster heads are marked, and the comprehensive score of the candidate cluster heads is calculated. ; The ratio of a node's remaining energy to its initial energy: ; in The remaining energy of node i. The initial maximum energy of the node.

[0051] Link reliability between nodes and candidate cluster heads: ; in, Let i be the signal power from node i to cluster head j. This represents the power of environmental noise.

[0052] Topological stability: ,in The number of neighbors of node i. Node lifespan Total network uptime Weighting coefficients For example, α can be set to 0.5 and β to 0.3.

[0053] In practical implementation, it can be set , , .

[0054] See Figure 3 Initially, node 0 is a custom cluster head, nodes 1, 3 and 4 are already joined nodes, and node 2 is a node to be joined. After calculation by the cluster head recommendation method of this embodiment, node 1 is determined to be the recommended cluster head, and nodes 0, 2, 3 and 4 are already joined nodes.

[0055] Specifically, in this embodiment of the invention, the candidate cluster heads are first quantitatively evaluated based on their energy state, network communication quality, and topology stability to determine the cluster heads. Then, when new nodes join subsequently, or during the calculation... or Cluster head upgrades can be performed whenever the conditions for cluster head upgrades are met, thereby maximizing the network's rapid deployment performance, improving beam resource utilization, and providing strong support for various missions such as final combat.

[0056] In practice, the cluster head election and dynamic update embodiments of the present invention dynamically update the cluster head role based on the node mobility characteristics in order to maintain network stability.

[0057] In other words, after the initial cluster head election is completed, the cluster head node broadcasts a control frame to notify the subgroup nodes. Then, stability maintenance is performed, which involves counting the number of node neighbors and calculating transmission energy consumption based on the current network topology to maximize the benefits. .if Use the cluster head election strategy to upgrade the cluster head and the backup cluster head; if The original planned cluster head will be maintained until the cluster head and backup cluster head upgrades are completed.

[0058] In addition, during the flight of the drone, the embodiments of the present invention can calculate the weight value of itself leaving the cluster in real time through the cluster head.

[0059] In this embodiment of the invention, the cluster head determines whether to leave the cluster weight calculation. ,in, To decouple from the cluster weight value, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. The flag for adjusting the quality of the task is set. If the cluster head actively leaves the cluster as needed for the task, then... Set to 1.

[0060] Specifically, in this embodiment of the invention, the cluster head calculates in real time whether the value is less than a critical value Q, which can be pre-planned through task configuration. When the cluster weight value is removed... If a sudden failure occurs, the cluster head will initiate the refactoring process. See details below. Figure 4 When a cluster head encounters sudden events such as strong interference, physical impact, or falling into water, rendering it unable to communicate and unable to inform other backup cluster heads that it has left the network, the backup cluster head will initiate the reconstruction process. See details below. Figure 5 .

[0061] See Figure 4 The cluster head initiation and reconstruction process specifically includes: S401. The cluster head calculates the link establishment and numbering status between each backup cluster head and other nodes within the cluster. During routine operations and link maintenance by backup cluster heads, the cluster head obtains the position, speed, and link establishment status of each node. Based on this information, it can calculate the weight value for each backup cluster head to become the nominated cluster head. ; S402. Cluster heads are ranked technically based on the weight of the number of chain-connected nodes and the ranking weight of the number; Specifically, in this embodiment of the invention, the weight value for each backup cluster head to become the nominated cluster head is calculated one by one. The clusters are sorted, and the calculation results are communicated to each backup cluster leader. After being communicated, the cluster leader is removed from the network. The cluster leader election metric function is... ;in, This is the number of backup cluster heads and the number of nodes that have already established chains with the cluster head. The total number of members in the cluster. The weight of the number of nodes in the established chain. This represents the number of nodes that have already established a chain. Sorting weights for node addresses This is the node address number.

[0062] Based on priority and There is an order of magnitude difference. .

[0063] S403. Determine the nodes to be promoted, reconfigure the network, and inform each backup cluster head of the results; Calculate the weight value of each backup cluster head for becoming the nominated cluster head. The nodes are sorted, and the node corresponding to the maximum value is the nominated cluster head. This message is then broadcast to all backup cluster heads. S404. After receiving the message, the standby cluster head confirms whether it is the nominated cluster head. Upon receiving the message, the standby cluster head determines whether it is the nominated cluster head. If so, it initiates a handshake with each member in the original cluster head's network management time slot after N seconds. If it is not the new cluster head, it waits for the new cluster head to initiate a handshake with it. Once each member responds to the handshake, the new cluster network is established.

[0064] S405. Initiate the network establishment process and shake hands with other members in the original cluster head's time slot; S406, Initiate the network construction process and wait for the new cluster head to connect with you.

[0065] See Figure 5 The backup cluster head startup and reconstruction process in this embodiment of the invention specifically includes: S501: Each backup cluster head fails to handshake with the cluster head for m consecutive times, resulting in communication failure. The cluster head is considered to have been passively removed from the network. When a cluster head encounters sudden events such as strong interference, physical impact, or falling into water that renders it unable to communicate, it cannot notify other backup cluster heads that it has left the network. In this case, the backup cluster heads will wait... This is the handshake link maintenance message that should have been sent.

[0066] S502, the backup cluster heads inform each other that the cluster head has lost contact (the cluster head information is not transmitted in the link maintenance information).

[0067] If all m handshakes fail, the backup cluster head will first notify other backup cluster heads through normal link maintenance messages that it has disconnected from the original cluster head, and then count whether other backup cluster heads have disconnected from the original cluster head. S503: No node, whether or not a chain has been established, can establish a chain with the cluster head. If all other nodes in the backup cluster head that have established links with the original cluster head lose their links, it is determined that the original cluster head has left the network, and the backup cluster head initiates reconstruction. If the location and speed information of the original cluster head can be obtained through other nodes, an attempt is made to restore the original cluster head to the network.

[0068] S504. Initiate the reconstruction process, collect statistics on the link establishment status between this node and other nodes in the cluster, remaining energy, link awareness, distance and number to the original cluster head node, and calculate its own weight value. .

[0069] In this embodiment of the invention, the autonomous cluster head recommendation comprehensive metric function is: , For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. Distance weight between the standby cluster head and the original failed cluster head. This refers to the distance between the backup cluster head and the original failed cluster head. To prevent the elected cluster head from failing or detaching again in the vicinity, It has a higher priority. , .

[0070] S505: Backup cluster heads send their respective weight values ​​to each other, with the highest weight value being the recommended cluster head. S506. Based on information from other nodes, determine the location and speed of the cluster head, and attempt to re-establish a link with the cluster head; If the location and speed information of the original cluster head are obtained from other nodes, then an attempt is made to restore the network using the original cluster head; S507. The alternate cluster head confirms whether it is the nominated cluster head; S508, Initiate the network establishment process and shake hands with other members in the original cluster head's time slot; S509, Initiate the network construction process and wait for the new cluster head to connect with you.

[0071] Accordingly, embodiments of the present invention provide a pre-planned cluster leader election device for rapid network construction of UAV swarm-oriented networks, see [link to relevant documentation]. Figure 6 The device includes: The first processing unit is used to construct the directional network topology of the UAV cluster based on the initial position coordinates, movement speed and direction parameters of each UAV node in the UAV cluster. Specifically, in this embodiment of the invention, the topology of the drone swarm is first constructed, and then the cluster head is selected based on the topology. Of course, this topology is updated in real time as the drone nodes move, so the cluster head also needs to be updated in real time.

[0072] The second processing unit is used to calculate the comprehensive energy value of each UAV node based on its neighboring UAV node information according to the directional network topology. Drone nodes with a comprehensive energy value greater than a preset energy threshold are selected as potential candidate cluster heads. The neighboring drone nodes of each drone node are determined based on a preset minimum distance principle. The distance variance is used to characterize the centrality of the cluster heads. For speed consistency, used to characterize drone nodes The speed difference with its neighboring drone nodes The horizontal angle is used to characterize the delay in establishing a communication connection between UAV nodes during directional antenna polling. Used to evaluate the overall performance of drone nodes in terms of location. Used to evaluate the overall performance of drone nodes in terms of speed consistency. Used to evaluate the overall performance of UAV nodes in terms of distribution azimuth angle; Specifically, in this embodiment of the invention, the comprehensive energy value of a node is calculated based on neighboring drone nodes, and drone nodes with comprehensive energy values ​​greater than a preset energy threshold are selected as potential candidate cluster heads. Then, candidate cluster heads are obtained by further filtering from these potential candidate cluster heads.

[0073] In specific implementation, the distance variance of the embodiments of the present invention Used to characterize the centrality of the cluster head Indicates drone node and drone nodes The Euclidean distance between them; Indicates drone node The average distance to the other drone nodes in the cluster; This represents the set of all drone nodes within the cluster; Speed ​​consistency of embodiments of the present invention Used for quantifying drone nodes Differences in speed with neighbors, including differences in direction and speed difference ,in, Represents a node The velocity vector; This represents the average velocity vector of neighboring nodes; Indicates consistency in direction; Indicates the difference in speed; Horizontal angle of the present invention This is used to characterize the delay in establishing a communication connection during directional antenna polling, where threshold is the horizontal angle threshold. , Represents a node and nodes The difference in horizontal coordinates; Represents a node and nodes The vertical coordinate difference; For nodes and nodes The horizontal angle.

[0074] in, Using drone nodes as cluster centers, this method is used to evaluate the overall performance of drone nodes in terms of location. Using drone nodes as cluster centers, this method is used to evaluate the overall performance of drone nodes in terms of speed consistency. A table is used to evaluate the overall performance of UAV nodes in terms of distribution azimuth, with UAV nodes serving as cluster centers. It should be noted that, in the embodiments of the present invention, a preset energy threshold and , and The values ​​are determined through experiments. Of course, in specific implementation, those skilled in the art can set them arbitrarily according to actual needs, but this invention will not discuss this in detail.

[0075] In specific implementation, the centrality evaluation in this embodiment of the invention is as follows: Smaller nodes are closer to the center of the network topology and have a stronger chance of being cluster head candidates. In specific implementation, the embodiments of the present invention determine neighboring nodes based on a preset minimum distance principle. That is, the distances between each node and its nearest connected node are added together, and the node with the smallest sum is taken as a cluster. The minimum distance, i.e. the sum of the distances between nodes, can be specifically set according to the number of nodes in the drone swarm, the number of clusters, and the specific control timeliness, etc. For example, it can be set to 20 meters, that is, the sum of the distances between all nodes in the cluster is 20 meters. Nodes exceeding this value will be assigned to other clusters. The third processing unit is used to eliminate redundant UAV nodes covering overlapping areas within the potential candidate cluster heads based on the single-beam coverage capability of the directional antenna, thereby obtaining candidate cluster heads. Specifically, in this embodiment of the invention, the beam coverage range is calculated based on the single-beam coverage capability of the directional antenna, and the beam pointing is controlled by adjusting the phase of the phased array antenna based on a preset beam avoidance strategy, so as to avoid multiple UAV nodes entering the network at the same time under the same beam, and the potential candidate cluster heads are screened to determine the candidate cluster heads.

[0076] The beam avoidance strategy achieves dynamic beam pointing by adjusting the phase of the phased array antenna. , For the signal wavelength, d represents the element spacing, N represents the number of elements, and d represents the element spacing.

[0077] The quantitative evaluation unit is used to quantitatively evaluate the energy state, network communication quality and topology stability of the candidate cluster heads, and selects the UAV node with the highest quantitative evaluation score as the cluster head and the UAV node with the second highest quantitative evaluation score as the backup cluster head. according to For each candidate cluster head, a quantitative evaluation of its energy state, network communication quality, and topology stability is performed. Remaining battery power Weighted by remaining battery power. For network communication quality, As a weight for network communication quality, For topological stability, These are the topological stability weights. It should be noted that each weight in the embodiments of this invention can be determined experimentally, and those skilled in the art can set them specifically; this invention does not impose specific limitations on this.

[0078] The monitoring unit is used to monitor the network topology status in real time during the flight of the drone swarm, and to trigger the cluster head upgrade when the cluster head replacement trigger condition is met.

[0079] In specific implementation, embodiments of the present invention calculate the current network topology and transmission energy consumption within the cluster based on the ground control terminal or the current cluster head during cluster flight. And determine whether the cluster head needs to be upgraded.

[0080] In addition, in specific implementation, those skilled in the art can also set various cluster head upgrade triggering conditions according to actual needs. For example, during the flight of a drone swarm, the probability of the cluster head leaving the swarm can be determined by the current cluster head. in, To decouple from the cluster weight value, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. Adjust the quality assignment flag for the task; if If the value is less than the critical value Q and the current cluster head is not faulty, then the weight values ​​of each backup cluster head are calculated based on the current cluster head to become the nominated cluster head. And determine the next cluster head; if If the value is less than the critical value Q and the current cluster is faulty, then each backup cluster head calculates its own weight value to become the nominated cluster head. And through weight values The order determines the next cluster head.

[0081] In other words, for cluster head upgrades, embodiments of the present invention can determine the upgrade based on different cluster head upgrade conditions, or based on the current network topology and transmission energy consumption within the cluster. ,if If the value exceeds a preset calculation threshold, a cluster head upgrade is triggered, or the probability of the cluster head leaving the cluster is determined by the cluster head itself. ,if If the value is less than the critical value Q, the cluster head upgrade is triggered. Of course, in specific implementation, those skilled in the art can also set other upgrade rules according to actual needs to ensure the accuracy of the cluster head, thereby achieving efficient self-control of the drone swarm and ultimately completing the predetermined mission objectives through the drone swarm.

[0082] In this embodiment of the invention, the weight value of the proposed cluster head for each backup cluster head is calculated based on the current cluster head, and the next cluster head is determined, including: Current cluster head based on Calculate the weight value for each backup cluster head to become the nominated cluster head. ,in, The total number of members in the cluster. Weights for the number of drone nodes in the established blockchain. Assign weights to the drone node addresses. This refers to the drone node address number; The calculated weight values Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head, and all alternate cluster heads are informed accordingly.

[0083] Specifically, in this embodiment of the invention, the weight value for each backup cluster head to become the nominated cluster head is calculated one by one by the cluster heads. According to priority and Order of magnitude difference After each calculation is completed, the results are sorted and communicated to each backup cluster head. Upon receiving the message, the backup cluster head determines whether it is the nominated cluster head. If so, it initiates a handshake with each member in the network management time slot of the original cluster head after N seconds. If it is not the new cluster head, it waits for the new cluster head to handshake with it. Once each member responds to the handshake, the new cluster network is established.

[0084] Furthermore, the embodiments of the present invention describe calculating the weight values ​​of each backup cluster head to become a nominated cluster head. And determine the next cluster head by sorting, including: each backup cluster head based on Calculate the weight value for each to become the head of the nominated cluster. And by weighting all backup cluster heads. Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head; Specifically, in this embodiment of the invention, each backup cluster head calculates its weight value for becoming a nominated cluster head based on a comprehensive metric function for autonomously nominated cluster heads. This data is then sent to other backup cluster heads, and the weight values ​​of all nodes are obtained through statistical analysis. And sorted, wherein the autonomously nominated cluster head comprehensive metric function is: ,in, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of nodes that have already established a chain. For the current node The ratio of the damaged value to the normal value of the link. Distance weight between the standby cluster head and the original failed cluster head. The standby cluster head determines its own weight value based on the distance between the standby cluster head and the original failed cluster head. After ranking first, it updates its role to cluster head. N seconds later, in the network management time slot of the original cluster head, it initiates a handshake with each member. After each member responds to the handshake, the new cluster network is established.

[0085] In general, the method described in the embodiments of the present invention can realize the pre-planning and dynamic recommendation of cluster heads and backup cluster heads in directional networks. By accurately recommending cluster heads, the network can be built quickly, thereby improving the utilization rate of beam resources.

[0086] In other words, the cluster head election device of the present invention can realize the pre-planning and dynamic election of cluster heads and backup cluster heads in directional networks, while improving the network's rapid deployment performance, thereby improving beam resource utilization.

[0087] Meanwhile, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-described methods for rapid network establishment of UAV swarm-oriented networks.

[0088] The relevant content of the device embodiment and storage medium embodiment of the present invention can be understood by referring to the method embodiment of the present invention, and will not be discussed in detail here.

[0089] Although preferred embodiments of the invention have been disclosed for illustrative purposes, those skilled in the art will recognize that various modifications, additions, and substitutions are possible, and therefore the scope of the invention should not be limited to the embodiments described above.

Claims

1. A pre-planned cluster head election method for fast network formation of a directional network of a UAV swarm, characterized in that, The method comprises: Based on the initial position coordinates, motion speed and direction parameters of each unmanned aerial vehicle node in the unmanned aerial vehicle cluster, a directional network topology structure of the unmanned aerial vehicle cluster is constructed; According to the directional network topology, each unmanned aerial vehicle node calculates its own comprehensive energy value based on neighbor unmanned aerial vehicle node information , and screens unmanned aerial vehicle nodes with a comprehensive energy value greater than a preset energy threshold as potential candidate cluster heads According to the single-beam coverage capability of the directional antenna, redundant unmanned aerial vehicle nodes in the overlapping coverage area of the potential candidate cluster head are excluded to obtain a candidate cluster head; The candidate cluster head is quantitatively evaluated in terms of energy state, network communication quality and topological stability, the unmanned aerial vehicle node with the highest quantitative evaluation score is taken as the cluster head, and the unmanned aerial vehicle node with the second highest quantitative evaluation score is taken as the backup cluster head; During the flight of the unmanned aerial vehicle cluster, the network topology state is monitored in real time, and the cluster head is upgraded when the cluster head replacement trigger condition is met. wherein the neighbor UAV nodes of each UAV node are determined based on a preset minimum distance principle, a distance variance for characterizing the centrality of the cluster head, a velocity consistency for characterizing the difference in velocity between the UAV node and its neighbor UAV nodes, a velocity consistency for characterizing the difference in velocity between the UAV node and its neighbor UAV nodes, a horizontal angle for characterizing the time delay in establishing a communication connection between the UAV nodes when polling by the directional antenna, a weight for evaluating the comprehensive performance of the UAV node in terms of the position, a weight for evaluating the comprehensive performance of the UAV node in terms of the velocity consistency, a weight for evaluating the comprehensive performance of the UAV node in terms of the distribution azimuth.

2. The method of claim 1, wherein, the distance variance is wherein, denotes the Euclidean distance between the UAV node and the UAV node ; denotes the average distance of the UAV node to the remaining UAV nodes within the cluster; denotes the set of all UAV nodes within the cluster; The speed consistency is , including direction difference and speed size difference , wherein, The speed vector of the unmanned aerial vehicle node ; The average speed vector of the neighbor unmanned aerial vehicle node, The direction difference and The speed difference is abnormal quantity; the horizontal included angle is wherein threshold is a horizontal included angle threshold, a UAV node a horizontal included angle between the UAV node and the UAV node , represents a horizontal coordinate difference between the UAV node and the UAV node ; represents a vertical coordinate difference between the UAV node and the UAV node .

3. The method of claim 1, wherein, According to the single-beam coverage capability of the directional antenna, redundant unmanned aerial vehicle nodes in the overlapping coverage area of the potential candidate cluster head are excluded to obtain a candidate cluster head, comprising: According to the single-beam coverage capability of the directional antenna, the beam coverage range is calculated, the beam pointing is controlled by adjusting the phased array antenna phase based on a preset beam avoidance strategy to avoid multiple unmanned aerial vehicle nodes entering the network at the same time under the same beam, and the potential candidate cluster head is screened to obtain the candidate cluster head.

4. The method according to any one of claims 1 to 3, characterized in that, The candidate cluster head is quantitatively evaluated in terms of energy state, network communication quality and topological stability, comprising: According to a quantitative evaluation of energy state, network communication quality and topology stability is performed for each candidate cluster head, wherein, is the residual energy, is the residual energy weight, is the network communication quality, is the network communication quality weight, is the topology stability, is the topology stability weight.

5. The method according to any one of claims 1-3, characterized in that, The real-time monitoring of the network topology state, and the triggering of the cluster head upgrade when the cluster head replacement trigger condition is met, comprising: based on The current network topology state and transmission energy consumption of the UAV cluster are calculated in real time, and whether the cluster head needs to be upgraded is judged according to the calculation result, and if so, the cluster head upgrade is triggered. wherein, is the number of neighbor UAV nodes of the UAV node i, is the average connectivity of the UAV nodes, is the transmission energy consumption between the UAV nodes, is the total residual energy of the UAV nodes, is the network connectivity weight, is the transmission energy consumption weight.

6. The method according to any one of claims 1 to 3, characterized in that, The real-time monitoring of the network topology state, and the triggering of the cluster head upgrade when the cluster head replacement trigger condition is met, comprising: Determine the probability of a cluster leaving the cluster by its cluster head. ,in, For the remaining energy weight, It is the ratio of energy consumed to total energy. For link-aware weights, This represents the number of drone nodes that have already established a blockchain. For the current drone node The ratio of the damaged value to the normal value of the link. The indicator for adjusting the quality of the task; If If the weight value of each backup cluster head is less than the threshold Q and the current cluster head is not failed, the weight value of each backup cluster head becoming the elected cluster head is calculated by the current cluster head And the next cluster head is determined; If If the current cluster is failed and the weight value of each backup cluster head is less than the threshold Q, each backup cluster head calculates its own weight value to become the elected cluster head And the next cluster head is determined by the weight value ranking.

7. The method of claim 6, wherein, calculating, by the current cluster head, a weight value for each backup cluster head to become a nominated cluster head and determining a next cluster head, comprising: The current cluster head calculates the weight value of each backup cluster head becoming the elected cluster head according to The weight value of each backup cluster head becoming the elected cluster head is calculated one by one Wherein, The total number of cluster members is, The weight of the number of unmanned aerial vehicle nodes in the built chain is, The weight of the unmanned aerial vehicle node address sorting is, The unmanned aerial vehicle node address number is; The calculated weight values are sorted, and the weight values of the highest backup cluster heads are selected as the next cluster heads, and each backup cluster head is informed.

8. The method of claim 7, wherein, The respective backup cluster head calculates a weight value of itself becoming the elected cluster head and determines the next cluster head through weight value sorting Each backup cluster head is based on Calculate the weight value for each to become the head of the nominated cluster. And by weighting all backup cluster heads. Sort by weight values The highest-ranking alternate cluster head becomes the next cluster head; wherein, is a remaining energy weight, is a ratio of consumed energy to total energy, is a ratio of a damaged value to a normal value of a link between the current and the UAV node a distance weight of the backup cluster head and the original faulty cluster head, is a distance of the backup cluster head and the original faulty cluster head.​ 9. A pre-planned cluster head election device for fast network formation of a UAV swarm-oriented directional network, characterized in that, The device comprises: A first processing unit configured to construct a directional network topology structure of the unmanned aerial vehicle cluster based on initial position coordinates, motion speed and direction parameters of each unmanned aerial vehicle node in the unmanned aerial vehicle cluster; a second processing unit configured to calculate, according to the directional network topology, a comprehensive energy value of each UAV node based on information of neighbor UAV nodes of the UAV node , and screen a UAV node with a comprehensive energy value greater than a preset energy threshold as a potential candidate cluster head, wherein the neighbor UAV nodes of each UAV node are determined based on a preset minimum distance principle, is a distance variance, used to represent centrality of the cluster head, is a speed consistency, used to represent difference in speed of the UAV node and its neighbor UAV nodes, is a horizontal angle, used to represent time delay of establishing a communication connection between the UAV nodes when polling by the directional antenna, used to evaluate comprehensive performance of the UAV node in terms of position, used to evaluate comprehensive performance of the UAV node in terms of speed consistency, used to evaluate comprehensive performance of the UAV node in terms of distribution azimuth. A third processing unit configured to exclude redundant unmanned aerial vehicle nodes in the overlapping coverage area of the potential candidate cluster head according to the single-beam coverage capability of the directional antenna to obtain a candidate cluster head; A quantitative evaluation unit configured to quantitatively evaluate the candidate cluster head in terms of energy state, network communication quality and topological stability, take the unmanned aerial vehicle node with the highest quantitative evaluation score as the cluster head, and take the unmanned aerial vehicle node with the second highest quantitative evaluation score as the backup cluster head; A monitoring unit configured to monitor the network topology state in real time during the flight of the unmanned aerial vehicle cluster, and trigger the cluster head upgrade when the cluster head replacement trigger condition is met.

10. A computer-readable storage medium, the storage medium storing a computer program, the program being executed by a processor to implement the pre-planned cluster head election method for fast network construction of a directional network of an unmanned aerial vehicle cluster according to any one of claims 1-8.

Images