An unmanned aerial vehicle self-organizing network routing method based on OODA loop service orchestration

Abstract

The application discloses a kind of unmanned aerial vehicle self-organizing network routing methods based on OODA ring service arrangement, the unmanned aerial vehicle self-organizing network routing method includes the following steps: S1: unmanned aerial vehicle node generates the traffic matrix of each OODA link according to task demand, and the communication time of each link service is modeled;S2: with all the service communication time L obtained by modeling as reference, unmanned aerial vehicle node shortens service communication time as optimization target, and preliminary path planning is carried out to service by service arrangement sub-algorithm;S3: by routing optimization algorithm, the path planning of service is carried out again, and the communication acceleration of unmanned aerial vehicle cluster under OODA system is completed.By this unmanned aerial vehicle self-organizing network routing method, the communication acceleration of unmanned aerial vehicle cluster under OODA system is completed, and the reliable communication of unmanned aerial vehicle network is guaranteed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of communication, and particularly relates to a UAV self-organizing network routing method based on OODA ring service arrangement. BACKGROUND

[0002] The OODA ring combat structure is the key to its effectiveness. The OODA ring refers to the intelligent combat system observation-judgment-decision-action process (Observe, Orient, Decide, Act, OODA), and the OODA ring operation process is as shown in Figure 1 Observation refers to the process in which each module in the combat system perceives the situation of both sides through radio frequency, vision, super vision, and acoustic perception. Judgment refers to the process in which the combat system makes a good-bad judgment on the battlefield situation based on intelligent algorithms. Decision-making refers to the process in which the combat system formulates an optimal scheme through intelligent task planning technology. Execution refers to the process in which each combat module executes the combat task according to the decision and returns the task completion result. In the traditional combat structure, a large number of links need to be completed by artificial means, and the commander needs to independently complete the collection of battlefield information when executing the combat task and make decisions based on a large amount of information. In high-intensity confrontation, it is inevitable to make a wrong judgment, and the efficiency of artificial information processing is low, which is easy to delay the opportunity in combat. The OODA ring combat structure greatly improves the automation and intelligence level of the combat system, overcomes the various disadvantages of the traditional combat structure, and effectively improves the effectiveness of the combat system.

[0003] The UAV cluster, as an important part of the OODA ring combat system, has increasing value in military operations. The UAV cluster network refers to a communication system in which a large number of low-cost UAVs guarantee task implementation through cooperative communication. In the UAV cluster OODA system confrontation, shortening the OODA cycle of the own side and making the cycle shorter than that of the enemy side can make the own UAV cluster have an advantage in confrontation. In order to improve the combat effectiveness of the UAV cluster and shorten the OODA cycle, the communication of the UAV network must be accelerated to guarantee reliable communication of the UAV network.

[0004] The unmanned aerial vehicle cluster under the OODA loop combat system is composed of several unmanned aerial vehicles, each of which is installed with a detection, judgment, decision or action load according to the combat task requirement, so as to perform a task in one or several links of the OODA cycle. When the unmanned aerial vehicle has a communication requirement, a service is generated, then a plurality of data packets carrying the service are generated, and all the data packets are transmitted to a destination node to complete the service. The unmanned aerial vehicle network adopts a distributed structure, and the positions of the unmanned aerial vehicle nodes in the network are equal. Each unmanned aerial vehicle node needs to independently collect network information and complete routing decision. The unmanned aerial vehicle moves at a high speed according to the task requirement, and the link of the network may be intermittently interrupted due to the movement of the unmanned aerial vehicle node, and the channel quality is not stable. The routing protocol commonly used in the unmanned aerial vehicle network is the OLSR protocol. Compared with the traditional AODV protocol, the OLSR protocol can better adapt to the dynamic environment of the unmanned aerial vehicle network. However, the OLSR protocol does not consider the multiplexing of the link in the transmission process, and the data packet may experience a long queuing delay in the transmission process. The OLSR protocol also does not measure the confidence degree of the link, which leads to the loss of the data packet in the transmission process and cannot guarantee the reliable communication of the unmanned aerial vehicle cluster in the combat process. In summary, the traditional OLSR routing protocol cannot meet the communication requirement of the unmanned aerial vehicle cluster under the OODA loop combat system.

[0005] In view of the problems of the routing protocol used by the unmanned aerial vehicle cluster under the existing OODA system:

[0006] (1) When planning the path of the service, the multiplexing of the link in the network is not considered, which easily causes link congestion, makes a large number of packets accumulate in the queue of each node on the congested link, the data packet thus experiences a long queuing delay, the unmanned aerial vehicle network needs a longer time to complete a service, and the OODA cycle of the unmanned aerial vehicle cluster is increased, which is not conducive to the struggle for the initiative of the unmanned aerial vehicle cluster in the battlefield.

[0007] (2) When planning the path of the service, the confidence degree of each link in the network is not measured. Due to the high-speed movement in the execution of the task, the link of the network may be intermittently interrupted due to the movement of the unmanned aerial vehicle node, and the channel quality is not stable. The data packet is easily lost in the transmission process, and the requirement of reliable communication of the unmanned aerial vehicle network cannot be met. SUMMARY

[0008] The purpose of the present application is to overcome the problems of the prior art, and disclose a kind of unmanned aerial vehicle self-organizing network routing method based on OODA ring service arrangement, which realizes the communication acceleration of unmanned aerial vehicle cluster under OODA system by the unmanned aerial vehicle self-organizing network routing method, and guarantees the reliable communication of unmanned aerial vehicle network.

[0009] The purpose of the present application is realized by the following technical scheme:

[0010] A routing method for UAV ad hoc networks based on OODA ring service orchestration, the routing method for UAV ad hoc networks includes the following steps:

[0011] S1: The UAV node generates the traffic matrix of each OODA link according to the task requirements, and models the communication time of each link's traffic.

[0012] S2: All business communication times obtained from modeling L For reference, the drone node aims to shorten the business communication time and uses a business orchestration sub-algorithm to perform preliminary path planning for the business.

[0013] S3: Through routing optimization algorithms, the service re-plans its path, thereby accelerating communication for drone clusters under the OODA system.

[0014] According to a preferred embodiment, in step S1, the UAV node perceives the network topology state through the interaction of control messages during network initialization, and generates the traffic matrix of each OODA link at the beginning of each link.

[0015] According to a preferred implementation, the communication time of each stage of the service is modeled in the following manner:

[0016] Let the network be the first i The estimated value of individual business communication time is , The formula is expressed as follows:

[0017]

[0018] in n Indicates business i The journey from the source node to the destination node takes a total of [number] steps. n Jump, Indicates the business process has gone through the first stage. j Stability of time-hop links, Indicates the business process has gone through the first stage. j The degree of reuse of time-hop links, For business i Business volume.

[0019] According to a preferred embodiment, the first is obtained through modeling. i Estimation of individual service communication time Based on the estimated communication time of any service, the communication time of all services is estimated, assuming that there are multiple services in the network. n If there are multiple services, then the estimated communication time for all services is: L , L The expression is as follows: .

[0020] According to a preferred embodiment, in step S2, after obtaining the traffic volume matrix, the UAV node runs the traffic orchestration sub-algorithm, sorts the traffic volume, performs preliminary path planning for the traffic in a prescribed order, and modifies the link weights of the topology graph based on the path planning results.

[0021] According to a preferred embodiment, the service orchestration sub-algorithm includes:

[0022] When no network traffic is generated, the drone nodes use interactive control messages to perceive the entire network topology and obtain stability information for each link. Then, each drone node sets the weight of each link in the entire network topology to the initial value.

[0023] The drone node generates a business matrix describing all services across the network based on the aggregated business information. All services are sorted in descending order of volume. After sorting, all services select paths according to the following process:

[0024] S21: Select the service with the largest weight value that has not yet undergone path planning, and obtain the optimal path for the service based on the current global topology using Dijkstra's algorithm;

[0025] S22: If the weight of the link traversed by the selected service is the initial value, then update the link weight to the weight value of the service; if the link weight is not the initial value, then add the weight of the service to the weight value of the link to obtain the new weight value of the link; and then calculate the new weight of the link traversed by the service and update the global topology.

[0026] S23: Determine whether path planning has been performed for all services. If not, return to step S21; otherwise, end the service orchestration sub-algorithm.

[0027] According to a preferred embodiment, in step S3, after completing the service orchestration sub-algorithm, the UAV node starts running the route optimization sub-algorithm, completes the re-path planning of the service through several rounds of iteration, and modifies the link weights of the topology graph so that the data packets can be transmitted according to the planned path.

[0028] According to a preferred embodiment, the routing optimization sub-algorithm includes the following steps:

[0029] S31: A new iteration begins, and the nodes sort the business in descending order based on the volume of business.

[0030] S32: Based on the current path planning for each service, obtain the estimated communication time for all services. Select the business with the largest business volume that has not yet been traversed in this iteration, and obtain a new path plan for the business based on the current global topology;

[0031] S33: Calculate the estimated communication time for all services based on the new path planning for this service. ,Will and contrast;

[0032] like Greater than If so, then the original route planning for that business will be maintained; if Less than This will cause the service to adopt a new path planning, and modify the weights of each link in the global topology according to the new path planning to obtain a new global topology;

[0033] S34: If all business processes have been traversed in this iteration, proceed to step S35; otherwise, return to step S32.

[0034] S35: If the path planning of one or more services is changed through this round of traversal, return to step S31 to start a new round of iteration; if the path planning of any service is not changed through this round of iteration, end the route optimization sub-algorithm.

[0035] The aforementioned main solution of the present invention and its various further alternative solutions can be freely combined to form multiple solutions, all of which are solutions that can be adopted and are claimed by the present invention. Those skilled in the art, after understanding the solution of the present invention, will realize that there are many combinations based on existing technology and common knowledge, all of which are technical solutions to be protected by the present invention, and will not be exhaustively listed here.

[0036] The beneficial effects of this invention are:

[0037] This invention proposes a routing method for UAV ad hoc networks based on OODA ring service orchestration. According to the service volume matrix of the current stage of the OODA ring, it optimizes the IP link weights of the distributed UAV ad hoc network through two stages: service orchestration and route optimization. This reduces the service communication time at each stage of the OODA ring, enabling the user to act first and gain battlefield initiative. The main benefits are:

[0038] (1) This invention models the communication time of UAV network services under the OODA system and obtains the estimated value of the communication time of all services in any link of OODA, providing a reference for the path planning of UAV nodes. UAV nodes can judge the quality of planning by comparing the estimated values ​​of service communication time under different path planning, and make effective adjustments to the path planning of services based on the evaluation results, thereby reducing the service communication time of each stage of the OODA ring.

[0039] (2) This invention introduces the concept of service orchestration into the routing of UAV ad hoc networks. Nodes make routing decisions based on the characteristics of services rather than the characteristics of data packets. Compared with traditional routing protocols, this algorithm, based on the concept of service orchestration, requires nodes to consider the reuse of traversed links when planning paths for services, avoiding link congestion, effectively shortening the queuing delay of data packets, reducing the service communication time at each stage of the OODA ring, accelerating the communication of UAV clusters under the OODA system, and enhancing the ability of UAV clusters under the OODA system to compete for initiative.

[0040] (3) The present invention requires the UAV to send control messages to obtain link stability information when it is idle. When the node makes path planning for the service, it takes into account the stability of the link, so that the data packets carrying the service can reach the destination node through a relatively stable link as much as possible. This avoids the loss of data packets caused by intermittent interruption of the UAV network link, improves the success rate of message transmission, and ensures reliable communication of the UAV network. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the OODA loop operation process;

[0042] Figure 2 This is a flowchart illustrating the service orchestration sub-algorithm in the UAV self-organizing network routing method of the present invention;

[0043] Figure 3 This is a flowchart illustrating the routing optimization sub-algorithm in the UAV self-organizing network routing method of the present invention;

[0044] Figure 4 This is a schematic diagram of a drone network under the OODA architecture;

[0045] Figure 5 It is a business volume matrix for the unmanned aerial vehicle (UAV) network reconnaissance process;

[0046] Figure 6 This is a diagram illustrating the weights of each link after the business orchestration sub-algorithm is executed;

[0047] Figure 7 It is the path planning result after the operation of the business orchestration sub-algorithm;

[0048] Figure 8 This is a diagram illustrating the route optimization process. Detailed Implementation

[0049] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features can be combined with each other. It should be observed that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0050] This invention discloses a routing method for UAV self-organizing networks based on OODA ring service orchestration. This UAV self-organizing network routing method is a distributed routing method, where each node in the network independently collects network information and makes routing decisions, adapting to the self-organizing architecture of UAV networks.

[0051] Preferably, the UAV self-organizing network routing method includes the following steps:

[0052] Step S1: The UAV node generates the traffic matrix of each OODA link according to the task requirements, and models the communication time of each link's traffic.

[0053] In step S1, the UAV node perceives the network topology status through the interaction of control messages during network initialization and generates the traffic matrix of each OODA link at the beginning of each link.

[0054] To shorten the OODA cycle of drone swarms, it is necessary to accelerate the communication of drone network services and reduce the communication time of services. Therefore, we model the communication time of drone network services.

[0055] First, we model the communication time of any service in the network. Let the successful transmission time of the first service in the network be... i The time required for each business is , The derivation formula is as follows:

[0056]

[0057] in For business i The moment when the last data packet is successfully transmitted to the destination node. For business iThe time when the first data packet is generated. According to the above formula, the time required to complete a service is determined by both the latency required for transmitting the data packets carrying the service and the success rate of data packet transmission. The longer the latency required for data packet transmission, the later the last data packet of the service will arrive at the destination node; the lower the success rate of data packet transmission, the more likely the source node will have to retransmit lost data packets, which will also... Increasing the latency of data packets can prolong the communication time of services. The latency of data packets reaching the destination node can be further divided into transmission latency, queuing latency, and propagation latency. The transmission and processing latency required for data packets to reach the destination node are determined by the number of hops required for the service to reach the destination node; the fewer the hops, the less time is required to transmit and process the data packets. The queuing latency required for data packet transmission is determined by the degree of multiplexing of services on each link; the higher the degree of multiplexing of a service through a link, the longer the queuing latency. The success rate of data packet transmission is determined by the stability of each link traversed by the service; the higher the stability of the link, the greater the probability that the data packet will be successfully transmitted to the destination node.

[0058] In summary, the communication time for any service can be estimated. Let the communication time of the first service in the network be... i The estimated value of individual business communication time is , The formula is expressed as follows:

[0059]

[0060] in, n Indicates business i The journey from the source node to the destination node takes a total of [number] steps. n Jump, Indicates the business process has gone through the first stage. j Stability of time-hop links, The larger the value, the higher the link stability, and vice versa. Indicates the business process has gone through the first stage. j The degree of reuse of time-hop links, The higher the value, the higher the degree of link reuse, and vice versa. For business i The greater the volume of business, the longer the communication time for the business.

[0061] in, The derivation method for the value is as follows: When there is no data service to transmit in the network, each node in the network will interact with its neighboring nodes through HELLO messages, and determine the stability of its link to the neighboring node based on the result of the interaction. Let the initial value of the stability of each link be . The node sends the first [unclear] to its neighboring nodes. k When sending HELLO messages, The value is iterated according to the following formula, where , r The larger the value, the more likely it is to represent The more sensitive it is to the stability of the link.

[0062] Considering that the ACK from the neighboring node is not immediately received by the node, when it is sent k The node that received the HELLO message after the first round received the message from its neighbor node. i Reply to the HELLO message, The value is iterated according to the following formula:

[0063]

[0064] By observing the exchange of HELLO and ACK messages between nodes during network idle periods, one can obtain information reflecting the stability of the links between nodes. value.

[0065] Reuse of services on the link The derivation method is as follows: if the first... k The business volume of each business is ,link j For the first k The link that a service passes through when it reaches the j-th hop. If at this time there are a total of n Each business (including business) k If all of them have passed through link j, then for the first... k For each service, the reuse status of link j for

[0066]

[0067] The larger the value, the more severe the reuse of the service through the j-th link for the k-th service, and the greater the queuing delay required to pass through that link.

[0068] In summary, the first result is obtained through modeling. i Estimation of individual service communication time Based on the estimated communication time of any service, the communication time of all services is estimated, assuming that there are multiple services in the network. n If there are multiple services, then the estimated communication time for all services is: L , L The expression is as follows: .

[0069] Step S2: Use the modeling results to determine the duration of all business communication. L For reference, the drone node aims to shorten the business communication time and uses a business orchestration sub-algorithm to perform preliminary path planning for the business.

[0070] In step S2, after obtaining the traffic volume matrix, the UAV node runs the traffic orchestration sub-algorithm, sorts the traffic volume, performs preliminary path planning for the traffic in the prescribed order, and modifies the link weights of the topology graph based on the path planning results.

[0071] Preferably, refer to Figure 2 As shown, the service orchestration sub-algorithm includes:

[0072] When no network traffic is generated, the drone nodes use interactive control messages to perceive the entire network topology and obtain stability information for each link. Then, each drone node sets the weight of each link in the entire network topology to the initial value.

[0073] The drone node generates a business matrix describing all services across the network based on the aggregated business information. All services are sorted in descending order of volume. After sorting, all services select paths according to the following process:

[0074] S21: Select the service with the largest weight value that has not yet undergone path planning, and obtain the optimal path for the service based on the current global topology using Dijkstra's algorithm;

[0075] S22: If the weight of the link traversed by the selected service is the initial value, then update the link weight to the weight value of the service; if the link weight is not the initial value, then add the weight of the service to the weight value of the link to obtain the new weight value of the link; and then calculate the new weight of the link traversed by the service and update the global topology.

[0076] S23: Determine whether path planning has been performed for all services. If not, return to step S21; otherwise, end the service orchestration sub-algorithm.

[0077] Step S3: Re-plan the service paths using a routing optimization algorithm to accelerate communication for the UAV swarm under the OODA system. The purpose of the routing optimization sub-algorithm is to further optimize each service path based on the service orchestration, thereby reducing the estimated communication time L for all services in this stage.

[0078] In step S3, after completing the service orchestration sub-algorithm, the UAV node starts running the route optimization sub-algorithm. Through several rounds of iteration, it completes the re-path planning of the service and modifies the link weights of the topology graph so that the data packets can be transmitted according to the planned path.

[0079] Preferably, refer to Figure 3 As shown, the route optimization sub-algorithm includes the following steps:

[0080] S31: A new iteration begins, and the nodes sort the business in descending order based on the volume of business.

[0081] S32: Based on the current path planning for each service, obtain the estimated communication time for all services. Select the business with the largest business volume that has not yet been traversed in this iteration, and obtain a new path plan for the business based on the current global topology;

[0082] S33: Calculate the estimated communication time for all services based on the new path planning for this service. ,Will and contrast;

[0083] like Greater than If so, then the original route planning for that business will be maintained; if Less than This will cause the service to adopt a new path planning, and modify the weights of each link in the global topology according to the new path planning to obtain a new global topology;

[0084] S34: If all business processes have been traversed in this iteration, proceed to step S35; otherwise, return to step S32.

[0085] S35: If the path planning of one or more services is changed through this round of traversal, return to step S31 to start a new round of iteration; if the path planning of any service is not changed through this round of iteration, end the route optimization sub-algorithm.

[0086] This invention proposes a routing method for UAV ad hoc networks based on OODA ring service orchestration. According to the service volume matrix of the current stage of the OODA ring, it optimizes the IP link weights of the distributed UAV ad hoc network through two stages: service orchestration and route optimization. This reduces the service communication time at each stage of the OODA ring, enabling the user to act first and gain battlefield initiative. The main benefits are:

[0087] (1) This invention models the communication time of UAV network services under the OODA system and obtains the estimated value of the communication time of all services in any link of OODA, providing a reference for the path planning of UAV nodes. UAV nodes can judge the quality of planning by comparing the estimated values ​​of service communication time under different path planning, and make effective adjustments to the path planning of services based on the evaluation results, thereby reducing the service communication time of each stage of the OODA ring.

[0088] (2) This invention introduces the concept of service orchestration into the routing of UAV ad hoc networks. Nodes make routing decisions based on the characteristics of services rather than the characteristics of data packets. Compared with traditional routing protocols, this algorithm, based on the concept of service orchestration, requires nodes to consider the reuse of traversed links when planning paths for services, avoiding link congestion, effectively shortening the queuing delay of data packets, reducing the service communication time at each stage of the OODA ring, accelerating the communication of UAV clusters under the OODA system, and enhancing the ability of UAV clusters under the OODA system to compete for initiative.

[0089] (3) The present invention requires the UAV to send control messages to obtain link stability information when it is idle. When the node makes path planning for the service, it takes into account the stability of the link, so that the data packets carrying the service can reach the destination node through a relatively stable link as much as possible. This avoids the loss of data packets caused by intermittent interruption of the UAV network link, improves the success rate of message transmission, and ensures reliable communication of the UAV network.

[0090] Application Cases

[0091] A drone network using this routing method, such as Figure 4 As shown, each drone node carries a payload as needed and performs tasks in the corresponding stage of the OODA ring. Node 1 in the diagram carries both reconnaissance and operational payloads, thus generating services in the reconnaissance and operational stages. We will demonstrate the algorithm's operation using the reconnaissance stage as an example. When each stage of the OODA ring begins operation, the network generates a service matrix for that stage. The service matrix for the reconnaissance stage is as follows: Figure 5 As shown.

[0092] After obtaining the traffic volume matrix, the node executes the traffic orchestration sub-algorithm. First, the node sets the weight of each link in the global topology to its initial value (1 in this example), and sorts all traffic in descending order of traffic volume based on the traffic matrix from the reconnaissance phase. Then, the node selects the traffic with the largest traffic volume that has not yet undergone path planning, and obtains the optimal path for this traffic using Dijkstra's algorithm based on the current global topology. Next, the node modifies the weights of the links traversed by the traffic according to the rules, and completes the path planning for all traffic according to the traffic orchestration sub-algorithm process. After the traffic orchestration sub-algorithm is completed, the weights of each link in the entire network topology are as follows: Figure 6 As shown. The path planning results for each business are as follows. Figure 7 As shown, 1->3 6 in business 1 represents a business in the business matrix where the source node is node 1, the destination node is node 3, and the business volume is 6. As shown in the figure, the planned path of this business is 1->2->3, that is, starting from node 1, passing through node 2, and reaching node 3.

[0093] The node then runs the routing optimization sub-algorithm. Based on the service matrix from the reconnaissance phase, the node sorts the services and calculates an estimate of the communication time for each service. For example, if the planned path for service 2 is 1->2->3, and the nodes calculate the stability of the link through the exchange of control messages. The values ​​are all (set up (1), according to the formula, we can get for:

[0094]

[0095] The node obtains all 8 services in the network. Then calculate the estimated value of all current business communication times. L old Calculated L old The value is 112. The node then selects the service with the highest traffic volume to replan the path; in this example, the node selects service 1. Based on the current topology, a new path is replanned for service 1. After planning, it is found that the new path for service 1 is the same as the original path. Therefore, the node continues to select services that have not yet been traversed in this round to replan the path; in this example, the node selects service 2. According to the algorithm's flow, the node sequentially selects services 2 and 3 to replan the path. It is found that the new path is the same as the original path, so the node continues to select service 4 for path planning. After Dijkstra's algorithm, the optimal path for service 4 in the current network topology is found to be 3->2->1, which is different from the previous path 3->4->7->1. At this point, the node recalculates the estimated communication time for all services based on the new path planning, obtaining... L new Calculated L new It is 110, compared to L old Smaller, indicating that the new path planning can reduce the expected communication time for all services, therefore the nodes update the weights of each link in the network topology according to the new path planning for service 4, as follows: Figure 8 As shown.

[0096] The node traverses all services according to the algorithm flow. Because the original path was modified in the first iteration, the node needs to start the second iteration. After several iterations, the path planning no longer changes, the route optimization sub-algorithm ends, and the node updates the routing table according to the global topology and completes the forwarding of data packets.

[0097] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A routing method for unmanned aerial vehicle (UAV) ad hoc networks based on OODA ring service orchestration, characterized in that, The unmanned aerial vehicle (UAV) self-organizing network routing method includes the following steps: S1: The UAV node generates the traffic matrix of each OODA link according to the task requirements, and models the communication time of each link's traffic. S2: All business communication times obtained from modeling L For reference, the drone node aims to shorten the business communication time and uses a business orchestration sub-algorithm to perform preliminary path planning for the business. S3: Re-plans the path for services through routing optimization algorithms to accelerate communication of drone clusters under the OODA system; In step S2, after obtaining the traffic volume matrix, the UAV node runs the traffic orchestration sub-algorithm, sorts the traffic volume, performs preliminary path planning for the traffic in the prescribed order, and modifies the link weights of the topology graph based on the path planning results. The business orchestration sub-algorithm includes: When no network traffic is generated, the drone nodes use interactive control messages to perceive the entire network topology and obtain stability information for each link. Then, each drone node sets the weight of each link in the entire network topology to the initial value. The drone node generates a business matrix describing all services across the network based on the aggregated business information. All services are sorted in descending order of volume. After sorting, all services select paths according to the following process: S21: Select the service with the largest weight value that has not yet undergone path planning, and obtain the optimal path for the service based on the current global topology using Dijkstra's algorithm; S22: If the weight of the link traversed by the selected service is the initial value, then update the link weight to the weight value of the service; if the link weight is not the initial value, then add the weight of the service to the weight value of the link to obtain the new weight value of the link; and then calculate the new weight of the link traversed by the service and update the global topology. S23: Determine whether all services have undergone path planning. If not, return to step S21; otherwise, end the service orchestration sub-algorithm. In step S3, after completing the service orchestration sub-algorithm, the UAV node starts running the route optimization sub-algorithm. Through several rounds of iteration, it completes the re-path planning of the service and modifies the link weights of the topology graph so that the data packets can be transmitted according to the planned path. The routing optimization sub-algorithm includes the following steps: S31: A new iteration begins, and the nodes sort the business in descending order based on the volume of business. S32: Based on the current path planning for each service, obtain the estimated communication time for all services. Select the business with the largest business volume that has not yet been traversed in this iteration, and obtain a new path plan for the business based on the current global topology; S33: Calculate the estimated communication time for all services based on the new path planning for this service. ,Will and contrast; like Greater than If so, then the original route planning for that business will be maintained; if Less than This will cause the service to adopt a new path planning, and modify the weights of each link in the global topology according to the new path planning to obtain a new global topology; S34: If all business processes have been traversed in this iteration, proceed to step S35; otherwise, return to step S32. S35: If the path planning of one or more services is changed through this round of traversal, return to step S31 to start a new round of iteration; if the path planning of any service is not changed through this round of iteration, end the route optimization sub-algorithm.

2. The UAV self-organizing network routing method as described in claim 1, characterized in that, In step S1, the UAV node perceives the network topology status through the interaction of control messages during network initialization and generates the traffic matrix of each OODA link at the beginning of each link.

3. The UAV self-organizing network routing method as described in claim 2, characterized in that, The communication time of each stage of the business is modeled in the following way: Let the network be the first i The estimated value of individual business communication time is , The formula is expressed as follows: in n Indicates business i The journey from the source node to the destination node takes a total of [number] steps. n Jump, Indicates the business process has gone through the first stage. j Stability of time-hop links, Indicates the business process has gone through the first stage. j The degree of reuse of time-hop links, For business i Business volume.

4. The UAV self-organizing network routing method as described in claim 3, characterized in that, The first was obtained through modeling. i Estimation of individual service communication time Based on the estimated communication time of any service, the communication time of all services is estimated, assuming that there are multiple services in the network. n If there are multiple services, then the estimated communication time for all services is: L , L The expression is as follows: .

Images