Low-delay unmanned aerial vehicle ad hoc network routing method

Abstract

The invention discloses a low-delay unmanned aerial vehicle ad hoc network routing method, which comprises the following steps that an unmanned aerial vehicle ad hoc network is initialized, and each unmanned aerial vehicle node detects a link between the unmanned aerial vehicle node and an adjacent node; each unmanned aerial vehicle node generates a local link information table and then calculates and forms a routing table, and the routing table is maintained and updated; the unmanned aerial vehicle node serves as a sending end to forward the service data frame; the unmanned aerial vehicle nodes serve as intermediate nodes to forward service data frames; and the unmanned aerial vehicle node serves as a receiving end to forward the service data frame. According to the low-delay unmanned aerial vehicle ad hoc network routing method provided by the invention, the forwarding method, the frame format and the like of the service data frame are optimized, the transmission delay of the service data frame is effectively reduced, the task requirements of high real-time performance and high collaboration are met, and the routing method is suitable for communication scenes of various unmanned aerial vehicle ad hoc networks.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) ad hoc networks, and particularly to a low-latency routing method for UAV ad hoc networks. Background Art

[0002] With the rapid development of UAVs in the military, civilian, and commercial fields, their tasks have become increasingly complex. A single UAV often cannot complete tasks well, so a large number of UAVs are needed to cooperate and coordinate with each other to complete tasks. However, for the high real-time and high coordination requirements of UAV ad hoc networks, the routing schemes in traditional networks can no longer meet the needs, and a low-latency scheme for UAV ad hoc networks needs to be designed.

[0003] Regarding the low-latency requirements of UAV ad hoc networks, protocol modification based on the MAC layer is an optimization direction. Xie Haibo et al. proposed a multi-hop Ad Hoc MAC layer protocol that combines routing layer information design, using protocol control frames to carry routing information, reducing the overhead of handshake signals. Simulations show that this protocol can reduce the point-to-point delay, but the average point-to-point delay is still greater than 10 milliseconds. Guan Junming et al. proposed a parameter CAM to measure the node access ability, and jointly carried out cross-layer design of the MAC layer and the network layer, and proposed a congestion-aware routing CAOR protocol. Simulations show that this protocol can reduce the point-to-point delay, but the average point-to-point delay is still in the millisecond range. Ma Minjue proposed a label routing protocol (MLLS) based on the wireless local area network standard protocol, using a cross-layer design method to implement the routing function in the MAC layer. However, since it uses an on-demand method for route discovery, the optimization effect of this method on the end-to-end delay is limited, and the average end-to-end delay is still in the millisecond range. Yang Guanxia et al. proposed a MANET security routing algorithm based on improved MAC and μTESLA, effectively combining data integrity and confidentiality protection, improving the security performance, but the average end-to-end delay of this method is still in the millisecond range.

[0004] Combined with the above research in the field of UAV ad hoc networks, the existing modification of the routing protocol based on the MAC layer mainly modifies the routing frame, and the optimization effect on the average end-to-end delay is not sufficient to meet the low-latency requirements of UAV ad hoc networks. There is still a large room for optimization for the low-latency scheme of UAV ad hoc networks. Summary of the Invention

[0005] In view of the above defects of the prior art, the technical problem to be solved by the present invention is that in the field of existing UAV ad-hoc networks, modifying the routing protocol based on the MAC layer mainly involves modifying the routing frame, and the optimization effect on the average end-to-end delay is insufficient to meet the low-latency requirements of UAV ad-hoc networks. The present invention provides a low-latency routing method for UAV ad-hoc networks, which optimizes the forwarding method and frame format of service data frames, effectively reducing the transmission delay of service data frames to meet the requirements of high-real-time and high-collaboration tasks, and is applicable to various communication scenarios of UAV ad-hoc networks.

[0006] To achieve the above object, the present invention provides a low-latency routing method for UAV ad-hoc networks, including the following steps:

[0007] S1, initialize the UAV ad-hoc network, and each UAV node detects the link between it and its adjacent nodes;

[0008] S2, each UAV node generates an adjacent node set, thereby generating a local link information table;

[0009] S3, according to the local link information table, each UAV node calculates and forms a routing table, and maintains and updates the routing table;

[0010] S4, when the UAV node is the sender, after caching and parsing the service data frame input from the network interface, query the routing table and forward the service data frame from the communication channel;

[0011] S5, when the UAV node is an intermediate node, after synchronizing the service data frame input from the communication channel, forward the service data frame from the communication channel through hardware table lookup;

[0012] S6, when the UAV node is the receiver, after synchronizing the service data frame input from the communication channel, forward the service data frame from the communication channel through hardware table lookup and cache parsing, and output the service data frame from the network interface.

[0013] Further, in S1, when initializing the UAV ad-hoc network, each UAV node detects the link between it and its adjacent nodes. Specifically, when initially establishing the UAV ad-hoc network, each UAV node periodically sends a neighbor relationship discovery packet for link detection. After receiving the neighbor relationship discovery packet, the UAV node replies with a neighbor relationship response packet, and obtains the relevant information of the link with the adjacent node through link detection.

[0014] Further, the neighbor relationship discovery packet includes the IP address of the local node. The IP address corresponding to each communication channel of the local node is different, and the last 8 bits are the communication channel number of the communication channel.

[0015] Furthermore, the neighbor relationship response packet includes the IP address of this node and the IP addresses of adjacent nodes. The IP address of the adjacent node corresponding to each communication channel is different, and the last 8 bits are the communication channel numbers of the communication channels of the adjacent nodes.

[0016] Furthermore, in S2, each UAV node generates a set of adjacent nodes, thereby generating a local link information table, specifically including that each UAV node generates a set of adjacent nodes and stores the link information from this node to its neighbor nodes; each UAV node exchanges the set of adjacent nodes to generate a local link information table.

[0017] Furthermore, in the set of adjacent nodes, when the link changes, the information corresponding to this link in the set of adjacent nodes is updated.

[0018] Furthermore, in S3, based on the local link information table, each UAV node calculates and forms a routing table, and maintains and updates the routing table, specifically including that based on the local link information table, each UAV node takes itself as the root to calculate the shortest path tree, thereby calculating the optimal path to each UAV node in the network and generating a routing table; each UAV node calculates the local routing table cyclically based on the local link information table; the local link information tables of each node are maintained and updated through keep-alive packets and keep-alive response packets, and the local routing tables of each node are maintained and updated.

[0019] Furthermore, in S4, when the UAV node is the sender, after caching and parsing the service data frame input from the network interface, it queries the routing table and forwards the service data frame from the communication channel, specifically including that when the UAV node is the sender, the user terminal obtains the MAC address of the UAV node through an arp packet; the user terminal sets the destination MAC address and the source MAC address, encapsulates the service data and sends it to the UAV node through the network interface; after caching, parsing and verifying the service data frame input from the network interface by the UAV node, it parses the destination IP address of the service data frame and queries the routing table, distributes it to the corresponding communication channel, modifies the destination MAC address and the source MAC address, re-encapsulates and outputs it from the communication channel.

[0020] Furthermore, in S5, when the UAV node is an intermediate node, after synchronizing the service data frame input from the communication channel, it forwards the service data frame from the communication channel through hardware table lookup, specifically including that when the UAV node is an intermediate node, after synchronizing the service data frame input from the communication channel by the UAV node, it does not need to cache the complete service data frame, but directly parses it, obtains the destination IP address from the destination MAC address and queries the routing table, distributes it to the corresponding communication channel, modifies the source MAC address, re-encapsulates and outputs it from the communication channel.

[0021] Further, in S6, when the drone node acts as the receiving end, after synchronizing the service data frames input through the communication channel, it performs hardware look-up table and caching parsing, and outputs the service data frames from the network interface. Specifically, when the drone node acts as the receiving end, the drone node obtains the MAC address of the user end through the ARP packet; after synchronizing the service data frames input through the communication channel, it does not need to cache the complete service data frames, but directly parses them, obtains the destination IP address from the target MAC address and queries the routing table, distributes them to the network interface arbitration competition, and then performs caching parsing and verification, modifies the target MAC address and the source MAC address, re-frames them and outputs them from the network interface.

[0022] Technical effects

[0023] A low-latency drone ad-hoc network routing method provided by the present invention optimizes the forwarding method and frame format of service data frames, and reduces the latency in the transmission stage of service data frames after route establishment.

[0024] The present invention separates the calculation and maintenance of the routing table from the synchronization, look-up table and framing of service data frames by combining software and hardware (the software includes the ARM core part, programmed in C language, and the hardware includes the FPGA chip part, programmed in Verilog language), makes full use of the characteristics of high hardware processing speed and stable processing speed to achieve low-latency synchronization, look-up table and framing, processes routing frames and service data frames in parallel, makes the routing method of the present invention have lower and more stable processing latency, and thus reduces the end-to-end latency.

[0025] Compared with traditional routers working in the Internet, a series of complex operations such as routing lookup, access control list matching, address resolution, and priority management need to be performed for the look-up table and forwarding of service data frames. When performing the look-up table operation in the present invention, only the node number in the destination IP address needs to be queried in the existing routing table to obtain the next-hop node number and the corresponding communication channel number, and they are directly transmitted to the communication channel through the communication channel number for direct forwarding, that is, the next-hop node number and the communication channel number are directly corresponding, which simplifies the look-up table forwarding process and reduces the processing latency and queuing latency of service data frames.

[0026] In addition, the routing method of the present invention adopts direct forwarding, does not need to cache the complete service data frames before parsing, and redefines the frame format of service data frames. The destination IP address of the service data frame can be obtained by parsing the target MAC address in the MAC frame header at the data link layer, and direct hardware look-up table and non-delay framing are performed, which reduces the processing latency compared with store-and-forward, and thus reduces the end-to-end latency.

[0027] The following will further illustrate the concept, specific structure and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features and effects of the present invention. Description of the Drawings

[0028] Figure 1 is a schematic flowchart of a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention;

[0029] Figure 2 is a path diagram of low-latency forwarding of service data frames in a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention;

[0030] Figure 3 is a frame format diagram of service data frames between drone nodes in a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention;

[0031] Figure 4 is a flowchart of low-latency forwarding of service data frames in a drone node of a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention;

[0032] Figure 5 is a relationship diagram between the number of drone nodes and the end-to-end delay in a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention;

[0033] Figure 6 is a schematic diagram of test results of a low-latency drone ad-hoc network routing method according to a preferred embodiment of the present invention. Detailed Embodiment

[0034] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clear, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0035] In the following description, specific details such as specific internal programs and technologies are proposed for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0036] As Figure 1 shown, the present invention provides a low-latency drone ad-hoc network routing method, including the following steps:

[0037] S1. Initialize the UAV ad hoc network, and each UAV node detects the links to its adjacent nodes.

[0038] Each UAV node has to detect the links to its adjacent nodes. Due to the uncertainty of wireless propagation, some links may be considered unidirectional. Therefore, all links must be authenticated bidirectionally to be considered available.

[0039] When initially establishing the UAV ad hoc network, each UAV node has to detect the links to its adjacent nodes. When there is no link connection relationship for each communication channel of each UAV node, it will periodically send neighbor discovery packets to detect adjacent nodes. Each UAV node periodically sends neighbor discovery packets for link detection. After receiving the neighbor discovery packet, the UAV node replies with a neighbor response packet. Through the interaction of the neighbor discovery packet and the neighbor response packet, relevant information about the link to the adjacent node can be obtained, including the node numbers and communication channel numbers of the two nodes, etc. Among them, the neighbor discovery packet includes the IP address of this node. Each communication channel corresponds to a different IP address of this node, and the last 8 bits are the communication channel number of this communication channel. The neighbor response packet includes the IP address of this node and the IP address of the adjacent node. Each communication channel corresponds to a different IP address of the adjacent node, and the last 8 bits are the communication channel number of the adjacent node's communication channel. Therefore, relevant information about the link to the adjacent node can be obtained through link detection, including the node numbers and communication channel numbers of the two nodes, etc.

[0040] S2. Each UAV node generates an adjacent node set, thereby generating a local link information table; specifically, each UAV node generates an adjacent node set and stores the link information from this node to its neighbor nodes; each UAV node exchanges the adjacent node sets to generate a local link information table. In the adjacent node set, when the link changes, update the information corresponding to this link in the adjacent node set. The local link information table stores the information of all links in the network known to this node. The UAV node can know the topological information of each UAV node and its adjacent nodes in the network through this table.

[0041] After receiving the neighbor response packet each time, maintain and update the adjacent node set. When the link changes, that is, when adding, modifying, or deleting adjacent nodes, update the corresponding part of the link in the adjacent node set according to the port number of this node corresponding to this link. Each UAV node exchanges the adjacent node sets to generate a local link information table. If the content in the local link information table changes, the UAV node will send a new topology control message to the adjacent nodes to update the relevant information in the local link information table of the adjacent nodes.

[0042] S3. According to the local link information table, each UAV node calculates and forms a routing table, and maintains and updates the routing table. Specifically, based on the local link information table, each UAV node takes itself as the root and calculates the shortest path tree, so as to calculate the optimal path to each UAV node in the network, and generates a routing table. Among them, based on the local link information table, each UAV node takes itself as the root and calculates the shortest path tree. The algorithm steps are as follows:

[0043] (1) Create a minimum heap with a size of V: V is the total number of UAV nodes in the local link information table, and each node in the minimum heap contains a node number and a weight value;

[0044] (2) Initialize the minimum heap: The weight value assigned to this UAV node is 0, and the weight value assigned to other nodes is INF (infinity);

[0045] (3) When the minimum heap is not empty, perform the following operations: ① Extract the node i with the minimum weight value from the minimum heap; ② For each adjacent node j of node i, check whether node j is in the minimum heap. If node j is in the minimum heap and the weight value is greater than the weight value from node i to node j plus the weight value of node i, then update the weight value of node j;

[0046] (4) When the minimum heap is empty, obtain the shortest path tree: The paths from this UAV node to each UAV node in the network on the shortest path tree are the optimal paths.

[0047] Each UAV node calculates the local routing table cyclically based on the local link information table; maintains and updates the local link information tables of each node through keep-alive packets and keep-alive response packets, and maintains and updates the local routing tables of each node. The neighbor relationship between adjacent UAV nodes is maintained through the periodic interaction of keep-alive packets and keep-alive response packets.

[0048] S4. When the UAV node is the sender, after caching and parsing the service data frame input from the network interface, it queries the routing table and forwards the service data frame from the communication channel. Specifically, when the UAV node is the sender, the user terminal obtains the MAC address of the UAV node through an arp packet; the user terminal sets the destination MAC address and the source MAC address, encapsulates the service data and sends it to the UAV node through the network interface; the UAV node caches, parses and verifies the service data frame input from the network interface, parses the destination IP address of the service data frame and queries the routing table, distributes it to the corresponding communication channel, modifies the destination MAC address and the source MAC address, re-encapsulates and outputs it from the communication channel.

[0049] Such as Figure 2The figure shows the path diagram for low-latency forwarding of service data frames. The service data frames need to be forwarded from Client 1 to Client n. The UAV nodes transmit service data frames through a communication channel, and the UAV nodes and clients transmit service data frames through network interfaces. The client obtains the MAC address of the UAV node through an ARP packet, and the UAV node obtains the MAC address of the client through an ARP packet.

[0050] When UAV Node 1 acts as the sender, Client 1 sets the destination MAC address to the MAC address of UAV Node 1, sets the source MAC address to the MAC address of Client 1, adds FCS at the end of the frame, encapsulates the service data, and sends it to UAV Node 1 through the network interface. After caching and parsing the service data frame input from the network interface, UAV Node 1 performs a checksum on the service data frame. If the checksum is unsuccessful, the frame is discarded; if the checksum is successful, the destination IP address of the service data frame is parsed out, the destination IP address is checked and the routing table is queried, and it is distributed to the corresponding communication channel. The last 24 bits of the destination MAC address are modified to the last 16 bits of the destination IP address and the 8-bit destination IP address checksum, the source MAC address is modified to the MAC address of UAV Node 1, FCS is added at the end of the frame, and it is re-encapsulated and output from the communication channel to the next UAV node.

[0051] S5. When the UAV node acts as an intermediate node, after synchronizing the service data frame input from the communication channel, it forwards the service data frame through hardware table lookup and from the communication channel; specifically including: when the UAV node acts as an intermediate node, after synchronizing the service data frame input from the communication channel, it does not need to cache the complete service data frame, but directly parses it, obtains the destination IP address from the destination MAC address and queries the routing table, distributes it to the corresponding communication channel, modifies the source MAC address, re-encapsulates it and outputs it from the communication channel to the next UAV node.

[0052] S6. When the UAV node acts as the receiver, after synchronizing the service data frame input from the communication channel, it performs hardware table lookup and cache parsing, and outputs the service data frame from the network interface. Specifically, when the UAV node acts as the receiver, the UAV node obtains the MAC address of the client through an ARP packet; after synchronizing the service data frame input from the communication channel, it does not need to cache the complete service data frame, directly parses it, obtains the destination IP address from the destination MAC address, checks the destination IP address and queries the routing table, distributes it to the network interface arbitration competition and then performs cache parsing and checksum, modifies the destination MAC address and the source MAC address, re-encapsulates it and outputs it from the network interface.

[0053] More specifically, when the drone node n acts as a receiver, after synchronizing the service data frames input through the communication channel, the drone node n does not need to cache the complete service data frames, but directly parses them. It obtains the destination IP address from the target MAC address, verifies the destination IP address and queries the routing table, distributes it after the network interface arbitration competition, then caches and parses it, and verifies the service data frames. If the verification fails, the frame is discarded; if the verification is successful, the target MAC address is modified to the MAC address of the user end n, the source MAC address is modified to the MAC address of the drone node n, an FCS is added at the end of the frame, and the frame is re-encapsulated and output from the network interface to the user end n.

[0054] The delay in the embodiments of the present invention refers to the end-to-end delay of the transmission of service data frames, including transmission delay, propagation delay, processing delay, and queuing delay. The present invention mainly reduces the end-to-end delay by reducing the processing delay. To reduce the end-to-end delay, the frame structure of the service data frames in the MAC layer is reconstructed to reduce the delay of parsing the service data frames to obtain the destination IP address, thereby reducing the processing delay and thus reducing the end-to-end delay. That is, some bits of the target MAC address of the service data frames are changed to some bits of the destination IP address and the check code of the destination IP address. Therefore, the destination IP address can be obtained after parsing the target MAC address. In the ad hoc network of drones in the embodiments of the present invention, each drone is both a router and a host, that is, each router is only connected to one host. Therefore, a node number is assigned to each drone, that is, the network number in the traditional network IP address, and the host number in the traditional network IP address is used as the interface number of the node, that is, the communication channel number.

[0055] The communication rate of the drone node for receiving and sending service data frames is the same. When the drone node receives the service data frames, it uses direct forwarding and does not need to cache the complete service data frames before parsing. That is, when the drone node receives the service data frames from the communication channel, after hardware synchronization in the physical layer, it can obtain the destination IP address of the service data frames by parsing the target MAC address in the MAC frame header of the data link layer, directly perform hardware table lookup and non-delay framing, and output from the communication channel, reducing the processing delay. The various functions of the service data frames are implemented based on a platform that combines software and hardware. The software part of the platform realizes high-speed routing calculation, maintenance, and update of the routing table, and the hardware part realizes hardware synchronization, hardware table lookup, and non-delay framing of the service data frames. Separating these functions makes the delay of synchronizing, looking up the table, and framing the service data frames more stable, not only realizing parallel processing to reduce the queuing delay but also reducing the processing delay of the service data frames.

[0056] Such as Figure 3The frame format of the service data frame between UAV nodes is shown as follows. In order to achieve low-latency transmission of the service data frame, the routing method of the present invention redefines the MAC frame for transmitting service data between UAV nodes. The first 16 bits of the IP address are the same, both set to 0x0A00. The last 24 bits of the destination MAC address in the frame header of the MAC frame are changed to the last 16 bits of the destination IP address and 8-bit destination IP address check code. By advancing the position of parsing the destination IP address in the service data frame, the latency of parsing the service data frame to obtain the destination IP address is reduced, that is, the processing latency is shortened.

[0057] As Figure 4 The flowchart of low-latency forwarding of the service data frame in the UAV node is shown as follows. Among them, the generation and processing of the routing frame are completed by the software part (ARM), and the processing of the service data frame is completed by the hardware part (FPGA). The service data frame is input from the communication channel and is directly forwarded. It is not necessary to cache the complete service data frame and then parse it. Instead, after hardware synchronization, the destination IP address of the service data frame can be obtained by parsing the destination MAC address in the MAC frame header at the data link layer. After directly performing hardware look-up table through the destination node number, framing and output can be performed without delay. By redefining the destination MAC address, the hardware synchronization time is shortened to the time of 48-bit data transmission. The hardware look-up table is implemented by FPGA, which is about 5 FPGA clock beats. The output of the framing without delay is about the time of 10-bit data transmission. In the traditional routing method, store-and-forward is required, and parsing can only be performed after caching the complete service data frame. The latency is mainly related to the frame length of the service data frame.

[0058] Compared with the CPU serial processing of service data, FPGA can process the service data of each communication channel at high speed in parallel. The latency of synchronizing, looking up tables, and framing the service data frame is stable and the processing rate is high. It not only realizes multi-channel parallel processing to reduce the queuing latency, but also reduces the processing latency of the service data frame.

[0059] When the communication rate between UAV nodes is 12.5 Mbps, the FPGA clock rate is 125 Mbps, and the frame length of the service data frame is 1000 bytes, the average point-to-point latency in the routing method proposed by the present invention is less than 10 μs, while the average point-to-point latency in the traditional routing method is in the millisecond order of magnitude. That is, the present invention greatly reduces the end-to-end latency through the look-up table forwarding method combining software and hardware.

[0060] For example, when the number of nodes in the UAV ad hoc network is 3, the UAVs communicate pairwise, the weight value from UAV node 1 to UAV node 2 is 1, the weight value from UAV node 1 to UAV node 3 is 3, and the weight value from UAV node 2 to UAV node 3 is 1, the specific steps of the above UAV ad hoc network routing method are as follows:

[0061] S1. Initialize the ad-hoc network of drones. Drone nodes 1, 2, and 3 respectively send neighbor discovery packets at regular intervals through communication channels 1 and 2 to detect adjacent nodes, and their IP addresses are 0x0A000101, 0x0A000102, 0x0A000201, 0x0A000202, 0x0A000301, and 0x0A000302. After each drone node receives the neighbor relationship response packet from an adjacent node, it can obtain the relevant information of the link with the adjacent node, including the node numbers and communication channel numbers of the two nodes, etc. That is, node 1 obtains the information that its channel 1 is connected to channel 1 of node 2 and its channel 2 is connected to channel 1 of node 3; node 2 obtains the information that its channel 1 is connected to channel 1 of node 1 and its channel 2 is connected to channel 2 of node 3; node 3 obtains the information that its channel 1 is connected to channel 2 of node 1 and its channel 2 is connected to channel 2 of node 2.

[0062] S2. After each drone node receives the neighbor relationship response packet, it stores the relevant information in the adjacent node set and maintains and updates the adjacent node set. Each drone node exchanges the adjacent node set to generate a local link information table and maintains and updates the local link information table. The local link information table of each drone node includes the following information: the information that channel 1 of node 1 is connected to channel 1 of node 2 and channel 2 of node 1 is connected to channel 1 of node 3, and channel 2 of node 2 is connected to channel 2 of node 3.

[0063] S3. According to the local link information table, each drone node calculates the shortest path tree with itself as the root, thereby calculating the optimal path to each drone node in the network, generating a routing table, and maintaining and updating the routing table. In this example, the shortest path trees of the 3 drone nodes are all as Figure 5 shown.

[0064] When drone node 1 needs to transmit a service data frame to drone node 3:

[0065] When the drone node 1 acts as the sender, the user terminal 1 sets the target MAC address as the MAC address 0x010101010101 of the drone node 1, sets the source MAC address as the MAC address 0xA1A1A1A1A1A1 of the user terminal 1, adds FCS at the end of the frame, encapsulates the service data and sends it to the drone node 1 through the network interface. After caching and parsing the service data frame input from the network interface by the drone node 1, it checks the service data frame, parses out the destination IP address 0x0A000300 of the service data frame, checks the destination IP address and queries the routing table, distributes it to the communication channel 1, modifies the last 24 bits of the target MAC address as the last 16 bits of the destination IP address and 8-bit destination IP address check code, modifies the source MAC address as the MAC address 0x010101010101 of the drone node 1, adds FCS at the end of the frame, re-encapsulates and outputs it from the communication channel 1 to the drone node 2.

[0066] S5. When the drone node 2 acts as an intermediate node, after synchronizing the service data frame input from the communication channel 1, it does not need to cache the complete service data frame, but directly parses it, obtains the destination IP address 0x0A000300 from the target MAC address and queries the routing table, distributes it to the communication channel 2, modifies the source MAC address as the MAC address 0x020202020202 of the drone node 2, re-encapsulates and outputs it from the communication channel 2 to the drone node 3.

[0067] S6. When the drone node 3 acts as the receiver, after synchronizing the service data frame input from the communication channel 2, it does not need to cache the complete service data frame, but directly parses it, obtains the destination IP address 0x0A000300 from the target MAC address, checks the destination IP address and queries the routing table, distributes it to the network interface arbitration competition and then caches and parses it, and checks the service data frame, modifies the target MAC address as the MAC address 0xC3C3C3C3C3C3 of the user terminal 3, modifies the source MAC address as the MAC address 0x030303030303 of the drone node 3, adds FCS at the end of the frame, re-encapsulates and outputs it from the network interface to the user terminal 3.

[0068] The above embodiments are actually tested. The routing method of this embodiment is mainly implemented based on the ZYNQ7020 core board, and the end-to-end delay in this routing method is tested through the Vivado 2020.1 software. The test results show that the average value of the end-to-end delay in this embodiment is 13.71729 μs. Similarly, the end-to-end delay of the drone ad hoc network with the number of nodes from 2 to 6 is tested, and the test results are as Figure 6As shown in the figure. The routing method proposed by the present invention reduces the average point-to-point delay to within 10 μs, and is basically proportional to the number of UAV nodes, which not only verifies that the present invention reduces the end-to-end delay of service data frame transmission, but also verifies the stability of the point-to-point delay of service data frames.

[0069] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations according to the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should fall within the protection scope determined by the claims.

Claims

1. A low-latency self-organizing network routing method for unmanned aerial vehicles, characterized in that, It includes the following steps: S1. Initialize the UAV ad-hoc network, and each UAV node detects the links with its adjacent nodes; S2. Each UAV node generates an adjacent node set, thereby generating a local link information table; S3. According to the local link information table, each UAV node calculates and forms a routing table, and maintains and updates the routing table; S4. As a sending end, the UAV node caches and parses the service data frame input from the network port, queries the routing table and forwards the service data frame from the communication channel; S5. As an intermediate node, the UAV node synchronizes the service data frame input from the communication channel, and forwards the service data frame from the communication channel through hardware table lookup; S6. As a receiving end, the UAV node synchronizes the service data frame input from the communication channel, and through hardware table lookup and cache parsing, outputs the service data frame from the network port.

2. The low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S1. Initialize the UAV ad-hoc network, and each UAV node detects the links with its adjacent nodes. Specifically, when initially establishing the UAV ad-hoc network, each UAV node periodically sends neighbor relationship discovery packets for link detection. After receiving the neighbor relationship discovery packet, the UAV node replies with a neighbor relationship response packet, and obtains the relevant information of the link with the adjacent node through link detection.

3. The low-latency drone ad-hoc network routing method according to claim 2, wherein, The neighbor relationship discovery packet includes the IP address of this node. Each communication channel corresponds to a different IP address of this node, and the last 8 bits are the communication channel number of this communication channel.

4. The low-latency drone ad-hoc network routing method according to claim 2, characterized in that, The neighbor relationship response packet includes the IP address of this node and the IP address of the adjacent node. Each communication channel corresponds to a different IP address of the adjacent node, and the last 8 bits are the communication channel number of the communication channel of the adjacent node.

5. A low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S2. Each UAV node generates an adjacent node set, thereby generating a local link information table, which specifically includes that each UAV node generates an adjacent node set and stores the link information from this node to the neighbor node; each UAV node exchanges the adjacent node sets to generate a local link information table.

6. A low-latency drone ad-hoc network routing method according to claim 5, characterized in that, In the adjacent node set, when the link changes, update the information corresponding to this link in the adjacent node set.

7. A low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S3. According to the local link information table, each UAV node calculates and forms a routing table, and maintains and updates the routing table, which specifically includes that based on the local link information table, each UAV node takes itself as the root and calculates the shortest path tree, thereby calculating the optimal path to each UAV node in the network and generating a routing table; each UAV node calculates the local routing table cyclically based on the local link information table; maintain and update the local link information tables of each node through keep-alive packets and keep-alive response packets, and maintain and update the local routing tables of each node.

8. A low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S4. When the UAV node acts as a sender, after caching and parsing the service data frame input from the network port, it queries the routing table and forwards the service data frame from the communication channel. Specifically, when the UAV node acts as a sender, the user terminal obtains the MAC address of the UAV node through an ARP packet; the user terminal sets the destination MAC address and the source MAC address, encapsulates the service data and sends it to the UAV node through the network port; after caching, parsing and verifying the service data frame input from the network port by the UAV node, it parses the destination IP address of the service data frame, queries the routing table, distributes it to the corresponding communication channel, modifies the destination MAC address and the source MAC address, re - encapsulates and outputs it from the communication channel.

9. The low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S5. When the UAV node acts as an intermediate node, after synchronizing the service data frame input from the communication channel, it forwards the service data frame through hardware table lookup from the communication channel. Specifically, when the UAV node acts as an intermediate node, after synchronizing the service data frame input from the communication channel, it does not need to cache the complete service data frame, but directly parses it, obtains the destination IP address from the destination MAC address, queries the routing table, distributes it to the corresponding communication channel, modifies the source MAC address, re - encapsulates and outputs it from the communication channel.

10. A low-latency drone ad-hoc network routing method according to claim 1, characterized in that, S6. When the UAV node acts as a receiver, after synchronizing the service data frame input from the communication channel, it forwards the service data frame through hardware table lookup and caching and parsing, and outputs the service data frame from the network port. Specifically, when the UAV node acts as a receiver, the UAV node obtains the MAC address of the user terminal through an ARP packet; after synchronizing the service data frame input from the communication channel, the UAV node does not need to cache the complete service data frame, directly parses it, obtains the destination IP address from the destination MAC address, queries the routing table, distributes it to the network port for arbitration competition and then performs caching, parsing and verification, modifies the destination MAC address and the source MAC address, re - encapsulates and outputs it from the network port.

Images