Self-adaptive routing strategy selection method applied to unmanned aerial vehicle cluster network

Abstract

The invention discloses a self-adaptive routing strategy selection method applied to an unmanned aerial vehicle cluster network, and belongs to the field of unmanned aerial vehicle cluster communication network routing planning. The method comprises a routing information generation process, a routing strategy generation process and a routing strategy execution process, the routing information generation process is to generate information such as a whole network link rate and a packet loss rate in an unmanned aerial vehicle cluster network based on an OLSR basic ad hoc network protocol; in the routing strategy generation process, a routing information sending strategy most suitable for the current network state is automatically selected based on the information transmission time and the information packet loss rate, the information delivery rate is improved, the network overhead is reduced, and the routing strategy execution process is based on different network states, so that the routing efficiency is improved. And adaptively selecting a routing path with the minimum time delay and a proper sending mode. According to the invention, in a high-dynamic and multi-rate unmanned aerial vehicle cluster communication environment, a proper routing sending strategy and a routing path can be actively selected, the information delivery rate is improved, and the network overhead is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of routing planning for unmanned aerial vehicle cluster communication networks, and in particular to an adaptive routing strategy selection method applied to unmanned aerial vehicle cluster networks. Background Art

[0002] At present, with the rapid development of drone equipment, its load-bearing capacity and flight endurance have also been rapidly improved. The academic and industrial circles have attached more and more importance to the research and application exploration of various aspects of drone cluster communication networks. With the expansion of the applicable scenarios of drone communication networks, some problems have gradually emerged. In the traditional single-drone communication guarantee system, the drone directly realizes the communication guarantee function for the ground communication coverage area, and the drone is used as an aerial base station to support emergency communications in a certain area. However, in complex aerial guarantee tasks, multiple drones will be needed in the air to form an aerial mobile network to ensure the communication of ground users. However, in a multi-drone communication network, in addition to considering the coordination of drone communication networks, it is also necessary to face the unevenness of the deployment coverage area of ​​the entire communication network, that is, the environment of each drone communication node in the drone network is different during the communication process, and the communication rate and network status of each node will also be very different. In addition, due to the high mobility characteristics of drones, the dynamic nature of drones in the air directly leads to intermittent connection relationships between neighbor links, which directly affects the normal transmission of information. These factors will make the data transmission efficiency of the drone communication network very low.

[0003] In recent years, with the application of drones in communications, more and more communication equipment has begun to be installed on drones to ensure communications in crowded areas, and drone flight control systems and communication technologies have become more mature. In order to better ensure ground communications, the optimization of routing and forwarding strategies in drone swarm networks has begun to be studied, further expanding the application scenarios of drones. Based on the above background, the selection of routing strategies for drone swarm networks is a key technical issue in optimizing drone communication networks and ensuring efficient data transmission in the network.

[0004] When drone communication nodes are performing tasks, there will be large differences in the geographical environment, electromagnetic environment, and node dynamics of different drone nodes. In actual communication tasks, a user on the ground usually needs to communicate with other nodes in the network. During the transmission of information packets from the source node to the destination node, they will pass through multiple aerial drone relay nodes. The communication status of each drone in the communication path will also be different. If the routing protocol is directly used to directly transmit data along the planned path without considering the real-time changes in the external situation environment, the reliability of information transmission may be greatly reduced, resulting in the loss of data packets. Therefore, a simple single-path shortest routing algorithm cannot be used to decide the final communication link. It is necessary to combine external real-time status information to select the optimal routing strategy, and transmit the calculated optimal routing strategy information to the networking system for reliable information transmission, to ensure the efficient operation of the network, and to maximize the benefits of network information transmission. Summary of the invention

[0005] In view of this, the present invention proposes an adaptive routing strategy selection method applied to a UAV cluster network, which can select the optimal routing strategy for UAVs in a complex dynamic environment.

[0006] The technical solution adopted by the present invention is:

[0007] An adaptive routing strategy selection method applied to a UAV cluster network, applied to a communication network with M UAV nodes, comprises the following steps:

[0008] Step 1: UAV node S measures the link speed of each neighbor node, broadcasts the network status to the entire network, and receives the network status of neighbor links; the network status includes: the switching threshold probabilities p1 and p2 of the network routing protocol, and the link rate v between any two neighbor nodes i and node j in the network i,j , p1>p2;

[0009] Step 2: UAV node S calculates the communication transmission time t between the kth UAV node and the k+1th UAV node in the network k,k+1 :

[0010]

[0011] Among them, Q is the size of the data packet transmitted by the drone node S;

[0012] Step 3, the drone node S uses a dynamic programming algorithm to continuously calculate twice the path with the minimum total transmission time from itself to the destination drone node D, where the first time generates the path L1 with the minimum total transmission time, and the second time generates the path L2 with the minimum total transmission time and not completely overlapping with L1;

[0013] Step 4: The drone node S is based on the historical information delivery rate p within the time range T. i,j , generate routing path strategy:

[0014] When p i,j When ≥p1, the single-path routing strategy is selected for sending, that is, the data is directly sent using path L1;

[0015] When p1≥p i,j When ≥p2, multipath routing strategy is selected for sending, that is, the same data packet is repeatedly sent in two paths, L1 and L2, to improve the reliability of information transmission;

[0016] When p2 ≥ p i,j When the data is received, the broadcast repeated sending strategy is selected, that is, the path is no longer searched from the routing table, but the data is directly broadcast to the outside. After each node receives the data, it no longer searches for the routing path, but directly broadcasts the data repeatedly until the data is repeated 3 times.

[0017] Furthermore, the specific method of step 3 is:

[0018] Step 301: Establish the node space of drone node S, including all nodes in the network, define an empty drone communication node sequence V, store the destination drone node D in the node sequence V; start the calculation from the destination drone node D, and find the node S that can directly communicate with node D with the shortest transmission time. m , and S m Add to the node sequence V;

[0019] Step 302: In the node space of the drone node S, remove the nodes in the node sequence V. For the last node S in the node sequence V, i , find the remaining nodes that can be connected to node S i Node S that communicates directly with the shortest transmission time k , S k Add to the node sequence V;

[0020] Step 303, repeat step 302 until the last node S in the node sequence V i It is able to communicate directly with node S. In this case, node S is added to the node sequence V and the node sequence V is output, which is the path L1 from node S to node D.

[0021] Step 304, re-establish the node space of drone node S, remove the nodes except node S and node D in path L1, and repeat steps 301 to 303. In this process, if no node space can be found that can be connected to the last node S in node sequence V, iIf the node is a node that can communicate directly with node S, add the nodes other than node S and node D in path L1 back to the node L1, and wait for a node that can communicate with node S. i After the node with direct communication and the shortest transmission time is found, the added nodes are removed again until node S is added to the node sequence V to generate path L2.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] 1. The present invention improves the efficiency of data transmission in the entire network and reduces the packet loss rate.

[0024] 2. The present invention improves the environmental adaptability of the entire network and ensures the adaptive generation of drone network topology routing in complex environments.

[0025] 3. The present invention designs a variety of routing path strategies and adaptively generates routing strategies, which can improve the reliability of information transmission and ensure the accuracy of ground user data transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention.

[0027] Figure 2 Schematic diagram of the communication routing path of the drone cluster in an embodiment of the present invention.

[0028] Figure 3 4 is a comparison diagram of the media occupancy time of different routes in an embodiment of the present invention.

[0029] Figure 4 This is a flow chart of routing strategy generation in an embodiment of the present invention.

[0030] Figure 5 The flowchart is a flow chart of selecting the optimal communication routing path in an embodiment of the present invention.

[0031] Figure 6 The figure is a flow chart of route set selection in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0033] An adaptive routing strategy selection method for unmanned aerial vehicle cluster networks includes the processes of routing information generation, routing strategy generation, routing strategy execution, etc. The routing information generation process is based on the OLSR basic self-organizing network protocol to generate information such as the full network link rate and packet loss rate in the unmanned aerial vehicle cluster network. The routing strategy generation process is based on the information transmission time and information packet loss rate to automatically select the routing information sending strategy that is most suitable for the current network state, improve the information delivery rate, and reduce network overhead. The routing strategy execution process is based on different network states, and adaptively selects the routing path with the smallest delay and the appropriate sending method.

[0034] In this method, the drone cluster calculates the communication status of all links in the network, that is, the communication rate, through the external network information known in advance (including network link rate, data packet size, network state switching threshold probability, etc.), and selects the two best paths according to the total communication transmission time. Figure 1 As shown in the figure, during the information transmission process, the state of the drone cluster network will change continuously. The network will count the information delivery rate of each node within a fixed time T, and will determine the routing strategy of the nodes in the network according to the change of the information delivery rate. It will also switch according to the specific strategy in the network and choose single-path routing transmission or multi-path routing transmission.

[0035] like Figure 1 As shown, the method comprises the following steps:

[0036] (1) Data input: The initial link communication time calculation module needs to obtain the communication rate v between any two nodes. i,j ,Finally, according to the communication rate, we can get the information transmission time between any two communicable nodes under the limited data packet size Q;

[0037] (2) Link communication time calculation: Based on the input communication rate and communication link status, the link communication time between any two points is generated:

[0038]

[0039] (3) Based on the minimum media time occupancy algorithm, two paths L1 and L2 with the shortest end-to-end time link in the entire network are generated, as follows: Figure 2 and 3 The specific method is:

[0040] Minimum media delay path calculation: Under the constraint of a fixed unit data packet size Q, calculate the time required for the data packet to pass between the kth node drone and the k+1th node drone, which is the communication transmission time t between the kth node drone and the k+1th node drone. k,k+1 :

[0041]

[0042] Minimum delay path selection: Assuming that there are M communication nodes in the communication network link, after knowing the communication transmission time from the kth node drone to the k+1th node drone, calculate the communication transmission time cost (t k,k+1 )for:

[0043]

[0044] Minimum delay path generation: After solving the information sending time of all routing paths, it can be seen that this problem is the TSP traveling salesman problem. The final problem is:

[0045] minT total

[0046] In the traditional method of solving the TSP problem, a dynamic programming algorithm can be used to solve the shortest path problem, that is, the path that occupies the least media time in the channel when sending information. Figure 5 As shown, the steps of the dynamic programming algorithm to solve the minimum media time path are as follows Figure 6 As shown, the details are as follows:

[0047] a) Define the drone communication node sequence V and store D in the sequence V. Starting from the end point of the fixed path, i.e. point D in the graph, find the node S that can directly communicate with node D with the shortest media occupancy time. m , and S m Add to sequence V.

[0048] b) In the node space, remove the nodes in the node sequence V and find a node S among the remaining nodes that can minimize the time consumed by the previous nodes, that is, the time in formula (1). k , and S k Add the node sequence to V.

[0049] c) Repeat step b until it is possible to communicate directly with the starting node S, and then add node S to the sequence V.

[0050] d) According to the order of nodes in sequence V, the data is sent in the order of the communication link path.

[0051] (4) Routing strategy generation: mainly includes generating routing path strategies based on information delivery rate and network dynamics of nodes in the network, such as Figure 4 As shown, specifically including:

[0052] When p i,j When ≥p1, a single-path routing path is selected for transmission;

[0053] When p1≥pi,j When ≥p2, select the broadcast routing path to send;

[0054] When p2 ≥ p i,j When selecting the broadcast repetition mechanism.

[0055] (5) Routing policy execution: Based on the routing path generated in step (3), the routing policy is executed.

[0056] Specifically include:

[0057] If single-path routing is used, the minimum-delay path is used for direct transmission;

[0058] If multi-path routing is used, the same data packet is repeatedly sent in two paths to improve the reliability of information transmission.

[0059] If broadcast transmission is performed, the path will no longer be searched in the routing table, and data will be directly broadcast to the outside. After each node receives the data, it will no longer search for a routing path, but will directly send the data repeatedly in a broadcast manner until A data packets are sent (A is set based on experience).

[0060] In summary, the present invention adaptively selects routing paths based on the information delivery rate of nodes and the dynamics of the network, can calculate the routing path strategy of the network based on the dynamics of the network and the information delivery rate, and adaptively selects the path strategy with the highest information delivery rate and the best information transmission effect, thereby adaptively adjusting and optimizing the routing strategy of the network based on the dynamic changes of the network. After selecting a suitable routing strategy, the present invention can realize single-hop, broadcast, and multiple broadcast transmissions respectively, ensure the efficient and reliable transmission of information data, and improve the delivery rate of information.

Claims

1. An adaptive routing strategy selection method applied to a drone cluster network, characterized in that: Applied to a communication network with M drone nodes, the following steps are included: Step 1: UAV node S measures the link speed of each neighbor node, broadcasts the network status to the entire network, and receives the network status of neighbor links; the network status includes: the switching threshold probabilities p1 and p2 of the network routing protocol, and the link rate v between any two neighbor nodes i and node j in the network i,j , p1>p2; Step 2: UAV node S calculates the communication transmission time t between the kth UAV node and the k+1th UAV node in the network k,k+1 : Among them, Q is the size of the data packet transmitted by the drone node S; Step 3, the drone node S uses a dynamic programming algorithm to continuously calculate twice the path with the minimum total transmission time from itself to the destination drone node D, where the first time generates the path L1 with the minimum total transmission time, and the second time generates the path L2 with the minimum total transmission time and not completely overlapping with L1; Step 4: The drone node S is based on the historical information delivery rate p within the time range T. i,j , generate routing path strategy: When p i,j When ≥p1, the single-path routing strategy is selected for sending, that is, the path L1 is directly used to send data; When p1≥p i,j When ≥p2, multipath routing strategy is selected for sending, that is, the same data packet is repeatedly sent in two paths, L1 and L2, to improve the reliability of information transmission; When p2 ≥ p i,j When the data is received, the broadcast repeated sending strategy is selected, that is, the path is no longer searched from the routing table, but the data is directly broadcast to the outside. After each node receives the data, it no longer searches for the routing path, but directly broadcasts the data repeatedly until the data is repeated 3 times.

2. The method for selecting an adaptive routing strategy for a drone cluster network according to claim 1, characterized in that: The specific method of step 3 is: Step 301, establish the node space of drone node S, including all nodes in the network, define an empty drone communication node sequence V, store the destination drone node D in the node sequence V; start the calculation from the destination drone node D, and find the node S that can communicate directly with node D with the shortest transmission time. m , and S m Add to the node sequence V; Step 302: In the node space of the drone node S, remove the nodes in the node sequence V. For the last node S in the node sequence V, i , find the remaining nodes that can be connected to node S i Node S that communicates directly with the shortest transmission time k , S k Add to the node sequence V; Step 303, repeat step 302 until the last node S in the node sequence V i It is able to communicate directly with node S. In this case, node S is added to the node sequence V and the node sequence V is output, which is the path L1 from node S to node D. Step 304, re-establish the node space of drone node S, remove the nodes except node S and node D in path L1, and repeat steps 301 to 303. In this process, if no node space can be found that can be connected to the last node S in node sequence V, i If the node is a node that can communicate directly with node S, add the nodes other than node S and node D in path L1 back to the node L1, and wait for a node that can communicate with node S. i After the node with direct communication and the shortest transmission time is found, the added nodes are removed again until node S is added to the node sequence V to generate path L2.

Images