A swarm-based UAV ad hoc networking method based on GNSS system time synchronization
By adopting a swarm-based UAV ad hoc networking method based on GNSS system time synchronization, the problems of frequent link interruptions, high transmission latency, and low channel utilization in UAV ad hoc networks are solved, and efficient and robust UAV ad hoc network communication is achieved.
Patent Information
- Application Number
- CN202310987390.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-07
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-08-07
AI Technical Summary
The ad hoc networks of unmanned aerial vehicles (UAVs) suffer from problems such as frequent link interruptions, high transmission latency, low channel utilization, and high protocol complexity, resulting in insufficient communication efficiency and robustness.
A swarm UAV self-organizing network method based on GNSS system time synchronization is adopted. By optimizing node routing through second pulse synchronization, MAC layer control access and routing layer protocols, the number of communication hops and routing nodes is reduced. By utilizing the global coverage of GNSS and high-precision second pulse synchronization, the nodes can be kept synchronized and communicate efficiently during rapid iteration.
It improves the communication efficiency and robustness of UAV ad hoc networks, reduces transmission latency, enhances system bandwidth, adapts to the requirements of rapid iteration of UAV ad hoc networks, and improves network access efficiency and network adaptability.
Smart Images

Figure CN117082650B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of unmanned aerial vehicle (UAV) communication technology, and in particular to a method for swarm UAV self-organizing network based on GNSS system time synchronization. Background Technology
[0002] Drones are trending towards miniaturization and swarm formation, and are being widely used in many areas of both military and civilian applications.
[0003] The networking structure of drone swarms is very similar to that of wireless ad hoc networks, both being decentralized and self-organizing. Drone ad hoc networks are a network architecture that combines wireless ad hoc network technology with drones. They possess characteristics such as independent networking, self-organization, and dynamic relay, playing a vital role in many fields, especially in tactical communications.
[0004] The following key issues typically exist in the practical application of drone self-organizing networks.
[0005] First, drones have limited energy and are highly dynamic, so they need to frequently enter and exit the network, which can easily cause rapid changes in topology and link interruption. Therefore, the network and MAC protocol for drone self-organizing networks need to have stronger tolerance for link interruption and rapid recovery capabilities.
[0006] Secondly, the high density of UAV networks means that a large number of nodes may communicate within a small area, making contention-based MAC protocols such as CSMA / CA unsuitable. This is because the high probability of transmission collisions leads to numerous retransmissions, which in turn increases transmission latency. TDMA-type protocols largely avoid transmission conflicts between nodes by dividing time slots, but fixed-allocation TDMA protocols have low channel utilization, while dynamic TDMA allocation protocols are complex and difficult to implement, resulting in many difficulties in the practical application of TDMA protocols.
[0007] Third, there is a lot of theoretical research on the MAC protocol of UAV ad hoc networks, but due to the complexity of the protocol, it places high demands on the robustness and real-time performance of hardware and software, and there is still relatively little research on the technical implementation of this. Summary of the Invention
[0008] This invention aims to at least partially solve one of the technical problems in related technologies. To this end, one objective of this invention is to propose a swarm UAV self-organizing network method based on GNSS system time synchronization. This method optimizes fixed network nodes or connection methods into variable nodes based on routing algorithms, finding the shortest root node communication route, reducing communication hop count or the number of routing nodes, and saving system bandwidth.
[0009] According to the present invention, a method for self-organizing UAV swarms based on GNSS system time synchronization is proposed, and the method steps are as follows:
[0010] S1: Second pulse synchronization method, which converts the 1PPS second pulse signal into a digital signal containing clock information and data information, and transmits it through the wireless communication module;
[0011] S2: MAC layer control access method, receives the digital signal sent in step S1, and after a given delay, completes the network access broadcast in sequence, filling the broadcast time frame-time slot-MAC data;
[0012] S3: Routing layer protocol, which takes the signal broadcast into the network completed in step S2, iterates it through the routing algorithm, establishes the routing table of each node, completes the establishment of the routing tree between each child node and the root node, and stores it;
[0013] S4: Ground control station, which numbers and assigns additional tags to nodes joining the network, exists as the root node or destination node of the ad hoc network, receives payload data and telemetry data, issues command and control instructions to nodes, and dynamically displays the routing path between each node through data tags.
[0014] Preferably, the steps of the second pulse synchronization method in step S1 are as follows:
[0015] As nodes, the self-organizing network drones all have the ability to synchronize second pulses, and are equipped with a GNSS receiver, an A / D signal conversion module, a D / A signal conversion module, a second pulse synchronization circuit and a wireless communication module.
[0016] The GNSS receiver outputs a 1PPS pulse per second.
[0017] The second pulse synchronization circuit is used to calibrate the 1PPS second pulse signal of the GNSS receiver and output the calibrated fine synchronization pulse.
[0018] The A / D signal conversion module receives radio frequency signals from adjacent nodes and converts them into digital signals;
[0019] The D / A signal conversion module converts digital signals into radio frequency signals, which carry the clock information and time frame data of this node.
[0020] The wireless communication module receives or transmits radio frequency signals.
[0021] Preferably, the second pulse synchronization circuit in step S1 includes a coarse synchronization circuit and a fine synchronization circuit. The coarse synchronization circuit calibrates the 1PPS second pulse signal of the GNSS receiver, and the fine synchronization circuit samples the clock information of adjacent nodes, performs an OR operation with the calibrated 1PPS second pulse, and outputs the fine synchronization pulse.
[0022] Preferably, the MAC layer control access method steps in step S2 are as follows:
[0023] S21: In an ad hoc network, any drone can act as a node. Any node sends a handshake packet at a predetermined period. Any node receives the handshake packets sent by its neighboring nodes and obtains information about its neighboring nodes.
[0024] S22: When any node sends, receives, or forwards information, it organizes the data from top to bottom in the form of time frame-time slot-MAC frame.
[0025] Preferably, in step S22, a time frame contains several time slots, and a time slot contains several MAC frames.
[0026] Preferably, time slot 0 carries specific messages, including a synchronization header; node number; measurement of the highest signal-to-noise ratio (SNR) signal of neighboring nodes; bandwidth negotiation; clock synchronization; and data type marking. Time slot N is the routing table and time frame data checksum, where N represents the last set of time slot numbers, and time slots 1 to N-1 are the data content.
[0027] Preferably, the routing layer protocol method steps in step S3 are as follows:
[0028] S31: Stores the communication routing table between this node and the ground control station;
[0029] S32: Process the received signal using a routing algorithm to obtain iterative data;
[0030] S33: Refresh the routing table every second based on the iterative data.
[0031] Preferably, the routing algorithm steps in step S32 are as follows:
[0032] S321: Any node receives and stores messages from all communicable nodes, and finds the root node route based on three parameters: shortest route, signal-to-noise ratio, and bandwidth.
[0033] S322: Route convergence is constrained by finding the root node. For routes that cannot reach the root node, the maximum route node is an 8-hop node.
[0034] S323: Any node stores all received routing tables, merges its own routing table, and sends it out within a 1-second communication cycle;
[0035] S324: Remove disconnected neighbor nodes based on signal-to-noise ratio information and update the routing table.
[0036] Preferably, the definition formula D of the routing tree i for:
[0037] Di = {Ri*0.5 + SNRi*0.4 + Bi*0.1}
[0038] Where i represents the ID number of any node in the ad hoc network; Ri represents the routing information between this node and all communicable nodes; SNRi represents the channel quality between this node and all communicable nodes; and Bi represents the bandwidth limit of neighboring nodes.
[0039] The beneficial effects of this invention are:
[0040] (1) The UAV ad hoc network is placed within a token ring network, with two improvements: First, any node within the token ring network can arbitrarily change its position based on a routing algorithm. Second, the collision probability time within the token ring network is deterministically delayed based on the arrival time of the second pulse, thus converting the collision detection time into a delay time. Optimizing the probabilistic acquisition of communication time slots into deterministic acquisition of communication time slots will increase system bandwidth. Optimizing fixed network nodes or connection methods into variable nodes based on a routing algorithm to find the shortest root node communication route reduces the number of communication hops or routing nodes, saving system bandwidth. This combination of improvement and reduction enhances the communication efficiency of the UAV ad hoc network and adapts to the real-world requirements of rapid iteration in UAV ad hoc networks.
[0041] (2) Due to the global coverage of GNSS navigation systems, the operational area of UAV ad hoc network communication is relatively small compared to the coverage area of GNSS navigation signals. Within this area, the second pulse error caused by ephemeris, ionospheric delay, tropospheric delay, multipath interference, and receiver noise is minimal, and is determined solely by the receiver's second pulse accuracy. By improving the receiver's second pulse accuracy and supplementing it with coarse and fine synchronization operations, the system's communication bandwidth can be further increased. At the same time, high-precision second pulses ensure that outgoing nodes remain synchronized with the ad hoc network after disconnection, thus avoiding handshake confirmation operations when joining the network and improving network entry efficiency.
[0042] (3) Improved robustness. The rapid movement of UAVs causes rapid changes in the ad hoc network routing tree, requiring each node to frequently iterate the routing algorithm. This invention uses a 1-second timeframe as the time limit, the shortest route root node as the algorithm target, and 8 hops as the number of route iterations. By imposing restrictions on time, the number of algorithm iterations, and the iteration target, it features the characteristic of prioritizing communication after obtaining the root node route, and reducing the weight of routes exceeding 8 hops, thus avoiding communication delays caused by network node congestion. Furthermore, by adjusting the weight ratios of the Di parameters, the network can be optimized in real time, improving its robustness to adapt to different operational tasks. Attached Figure Description
[0043] In the attached diagram:
[0044] Figure 1 This is a flowchart illustrating a method for self-organizing UAV swarms based on GNSS system time synchronization proposed in this invention.
[0045] Figure 2 This is a block diagram and module configuration of the second pulse synchronization principle proposed in this invention;
[0046] Figure 3 This is a schematic diagram illustrating the MAC frame, time slot, time frame construction method, and their interrelationships proposed in this invention.
[0047] Figure 4 This is a schematic diagram of the basic data reuse relationship of the routing protocol proposed in this invention;
[0048] Figure 5 This is a schematic diagram illustrating how the routing tree generation protocol proposed in this invention is determined by the Di value stored in the node. Detailed Implementation
[0049] Reference Figure 1 A method for self-organizing UAV swarms based on GNSS system time synchronization is described below:
[0050] S1: Second pulse synchronization method, which converts the 1PPS second pulse signal into a digital signal containing clock information and data information, and transmits it through the wireless communication module;
[0051] S2: MAC layer control access method, receives the digital signal sent in step S1, and after a given delay, completes the network access broadcast in sequence, filling the broadcast time frame-time slot-MAC data;
[0052] S3: Routing layer protocol, which takes the signal broadcast into the network completed in step S2, iterates it through the routing algorithm, establishes the routing table of each node, completes the establishment of the routing tree between each child node and the root node, and stores it;
[0053] S4: Ground control station, which numbers and assigns additional tags to nodes joining the network, exists as the root node or destination node of the ad hoc network, receives payload data and telemetry data, issues command and control instructions to nodes, and dynamically displays the routing path between each node through data tags.
[0054] Reference Figure 2 The steps of the second pulse synchronization method in step S1 are as follows:
[0055] As nodes, the self-organizing network drones all have the ability to synchronize second pulses, and are equipped with a GNSS receiver, an A / D signal conversion module, a D / A signal conversion module, a second pulse synchronization circuit and a wireless communication module.
[0056] The GNSS receiver outputs a 1PPS pulse per second.
[0057] The second pulse synchronization circuit is used to calibrate the 1PPS second pulse signal of the GNSS receiver and output the calibrated fine synchronization pulse.
[0058] The A / D signal conversion module receives radio frequency signals from adjacent nodes and converts them into digital signals;
[0059] The D / A signal conversion module converts digital signals into radio frequency signals, which carry the clock information and time frame data of this node.
[0060] The wireless communication module receives or transmits radio frequency signals.
[0061] In step S2, the second pulse synchronization circuit includes a coarse synchronization circuit and a fine synchronization circuit. The coarse synchronization circuit calibrates the 1PPS second pulse signal of the GNSS receiver. The fine synchronization circuit samples the clock information of adjacent nodes, performs an OR operation with the calibrated 1PPS second pulse, and outputs the fine synchronization pulse.
[0062] Reference Figure 3 The steps of the MAC layer control access method in step S2 are as follows:
[0063] S21: In an ad hoc network, any drone can act as a node. Any node sends a handshake packet at a predetermined period. Any node receives the handshake packets sent by its neighboring nodes and obtains information about its neighboring nodes.
[0064] S22: When any node sends, receives, or forwards information, it organizes the data from top to bottom in the form of time frame-time slot-MAC frame.
[0065] In step S22, a time frame contains several time slots, and a time slot contains several MAC frames.
[0066] Slot 0 carries specific messages, including a synchronization header; node number; measurement of the highest signal-to-noise ratio (SNR) signal of neighboring nodes; bandwidth negotiation; clock synchronization; and data type marking. Slot N contains the routing table and time frame data checksum, where N represents the last set of time slot numbers. Slots 1 to N-1 contain the data content.
[0067] Reference Figure 4 The steps of the routing layer protocol in step S3 are as follows:
[0068] S31: Stores the communication routing table between this node and the ground control station;
[0069] S32: Process the received signal using a routing algorithm to obtain iterative data;
[0070] S33: Refresh the routing table every second based on the iterative data.
[0071] Preferably, the routing algorithm steps in step S32 are as follows:
[0072] S321: Any node receives and stores messages from all communicable nodes, and finds the root node route based on three parameters: shortest route, signal-to-noise ratio, and bandwidth.
[0073] S322: Route convergence is constrained by finding the root node. For routes that cannot reach the root node, the maximum route node is an 8-hop node.
[0074] S323: Any node stores all received routing tables, merges its own routing table, and sends it out within a 1-second communication cycle;
[0075] S324: Remove disconnected neighbor nodes based on signal-to-noise ratio information and update the routing table;
[0076] The definition formula for a routing tree, D i for:
[0077] Di = {Ri*0.5 + SNRi*0.4 + Bi*0.1}
[0078] Where i represents the ID number of any node in the ad hoc network; Ri represents the routing information between this node and all communicable nodes; SNRi represents the channel quality between this node and all communicable nodes; and Bi represents the bandwidth limit of neighboring nodes.
[0079] The calculation process is as follows:
[0080] (1) According to the preset ID number (0-N) in size order, the root node and nodes 1-8, in turn, broadcast the message after a preset delay based on the arrival time of the synchronization second pulse. For example, node 1 delays by 1µs after the arrival of the precise synchronization second pulse, and node 8 delays by 8µs after the arrival of the precise synchronization second pulse;
[0081] (2) When any node broadcasts a message, other nodes listen for and receive the message, and calculate Di. SNRi is calculated from the physical layer RSSI (Received Signal Strength). Bi is given by the "bandwidth" value of frame 0-slot 0-MAC0 when the broadcast message is received.
[0082] (3) The calculation of Ri is divided into two steps: First, listen to the broadcast messages of all nodes within one PPS pulse period, each node obtains the broadcast messages of the most neighboring nodes and stores them; Second, in the second PPS pulse period, each node broadcasts the connection relationship between itself and its neighboring nodes, and each node stores the connection relationship between other nodes and their neighbors; After repeating for 8 PPS pulse periods, the original routing table with a maximum of 8 hops is completed; Finally, the weighted search algorithm is used to complete the calculation of Di.
[0083] (4) Since most of the tasks performed by the UAV ad hoc network are sending messages to the ground control station, i.e., the root node, or from the root node to other nodes, the convergence condition of the routing algorithm is: find the route from the node with the minimum number of hops to the root node; the number of hops from node to node is less than 8. After calculating the route according to the constraints, it is stored in the routing table.
[0084] (5) Example, with Figure 5 For example.
[0085] During the first PPS communication cycle, each node obtains the connection relationships of its neighboring nodes, as shown in the table below. Each node stores its neighboring nodes in its routing table and broadcasts them in the next communication cycle.
[0086]
[0087] Taking node 1 as an example, in the second PPS communication cycle, it obtains neighboring nodes 2 / 3 / 4 / 5 / 7 / 8. At this time, node 1 can search using the Di formula, the condition of which is to obtain the root node route, or a route with a maximum hop count < 8. See the table below. Starting from node 1, the root node can be reached via either node 3 or node 4. Since the SNR of node 1 ←→ node 4 is greater than that of node 1 ←→ node 3, node 1's routing table prioritizes forwarding from node 4 to the root node, and secondly, forwarding from node 3 to the root node.
[0088]
[0089] The above process is repeated 8 times to form Figure 5 Routing connection relationships.
[0090] Network entry and network exit. For example... Figure 5 As shown, since node 2 flies away from node 1 and communication is eventually interrupted, node 2 is deleted from all routing tables in node 1. Since node 8 flies closer to node 1, node 8 is added to node 1's routing table. Node 2, which leaves the network, still maintains the routing table of the entire ad hoc network. Once it establishes a communication connection with a node in the ad hoc network, it can obtain the entire network route and obtain the optimal routing table through a routing algorithm. Node 8, which enters the network, has an enhanced SNR and iterates through node 1's routing table using an algorithm.
Claims
1. A method for self-organizing UAV swarms based on GNSS system time synchronization, characterized in that, The steps are as follows: S1: Second pulse synchronization method, which converts the 1PPS second pulse signal into a digital signal containing clock information and data information, and transmits it through the wireless communication module; S2: MAC layer control access method, receives the digital signal sent in step S1, and after a given delay, completes the network access broadcast in sequence, filling the broadcast time frame one time slot - MAC data; S3: Routing layer protocol, which takes the signal broadcast into the network completed in step S2, iterates it through the routing algorithm, establishes the routing table of each node, completes the establishment of the routing tree between each child node and the root node, and stores it; S4: Ground control station, which numbers and assigns additional tags to nodes joining the network, exists as the root node or destination node of the ad hoc network, receives payload data and telemetry data, issues command and control instructions to nodes, and dynamically displays the routing path between each node through data tags.
2. The method for swarm UAV self-organizing network based on GNSS system time synchronization according to claim 1, characterized in that, The steps of the second pulse synchronization method in step S1 are as follows: As nodes, the self-organizing network drones all have the ability to synchronize second pulses, and are equipped with a GNSS receiver, an A / D signal conversion module, a D / A signal conversion module, a second pulse synchronization circuit and a wireless communication module. The GNSS receiver outputs a 1PPS pulse per second. The second pulse synchronization circuit is used to calibrate the 1PPS second pulse signal of the GNSS receiver and output the calibrated fine synchronization pulse. The A / D signal conversion module receives radio frequency signals from adjacent nodes and converts them into digital signals; The D / A signal conversion module converts digital signals into radio frequency signals, which carry the clock information and time frame data of this node. The wireless communication module receives or transmits radio frequency signals.
3. The method for swarm UAV self-organizing network based on GNSS system time synchronization according to claim 2, characterized in that: In step S1, the second pulse synchronization circuit includes a coarse synchronization circuit and a fine synchronization circuit. The coarse synchronization circuit calibrates the 1PPS second pulse signal of the GNSS receiver, and the fine synchronization circuit samples the clock information of adjacent nodes, performs an OR operation with the calibrated 1PPS second pulse, and outputs the fine synchronization pulse.
4. The method for swarm UAV self-organizing network based on GNSS system time synchronization according to claim 1, characterized in that, The steps of the MAC layer control access method in step S2 are as follows: S21: In an ad hoc network, any drone can act as a node. Any node sends a handshake packet at a predetermined period. Any node receives the handshake packets sent by its neighboring nodes and obtains information about its neighboring nodes. S22: When any node sends, receives, or forwards information, it organizes the data from top to bottom in the form of time frame-time slot-MAC frame.
5. The method for swarm UAV self-organizing network based on GNSS system time synchronization according to claim 4, characterized in that: In step S22, a time frame contains several time slots, and a time slot contains several MAC frames.
6. A method for swarm UAV self-organizing networking based on GNSS system time synchronization according to claim 5, characterized in that, Slot 0 carries specific messages, including a synchronization header; node number; measurement of the highest signal-to-noise ratio (SNR) signal of a neighboring node; bandwidth negotiation; and clock synchronization. Data type marker, time slot N is the routing table and time frame data checksum, where N represents the last group of time slot numbers, and time slots 1 to N-1 are the data content.
7. The method for swarm UAV self-organizing network based on GNSS system time synchronization according to claim 1, characterized in that, The steps of the routing layer protocol in step S3 are as follows: S31: Stores the communication routing table between this node and the ground control station; S32: Process the received signal using a routing algorithm to obtain iterative data; S33: Refresh the routing table every second based on the iterative data.
8. A method for swarm UAV self-organizing networking based on GNSS system time synchronization according to claim 7, characterized in that, The routing algorithm steps in step S32 are as follows: S321: Any node receives and stores messages from all communicable nodes, and finds the root node route based on three parameters: shortest route, signal-to-noise ratio, and bandwidth. S322: Route convergence is constrained by finding the root node. For routes that cannot reach the root node, the maximum route node is an 8-hop node. S323: Any node stores all received routing tables, merges its own routing table, and sends it out within a 1-second communication cycle; S324: Remove disconnected neighbor nodes based on signal-to-noise ratio information and update the routing table.
9. A method for swarm UAV self-organizing networking based on GNSS system time synchronization according to claim 8, characterized in that, The definition formula for a routing tree, D i for: Di = {Ri*0.5 + SNRi*0.4 + Bi*0.1} Where i represents the ID number of any node in the ad hoc network; Ri represents the routing information between this node and all communicable nodes; SNRi represents the channel quality between this node and all communicable nodes; and Bi represents the bandwidth limit of neighboring nodes.
Citation Information
Patent Citations
Ad hoc network communication system, method and frame structure based on lightweight clock synchronization mechanism
CN114885413A
Clock synchronization method for unmanned aerial vehicle cluster
CN115996459A
