A high-dynamic flight ad hoc network routing protocol construction method oriented to link robustness

Abstract

The application belongs to the technical field of flight ad hoc network communication, and discloses a high-dynamic flight ad hoc network routing protocol construction method for link robustness. In view of the problems of frequent disconnection and unstable transmission of the high-dynamic flight ad hoc network link, the application provides a high-dynamic flight ad hoc network routing protocol construction method for link robustness, which fuses the received signal strength (RSSI) and the link lifetime (Link Lifetime) by normalization weighting to construct a composite link robustness index. The index is combined to optimize MPR selection and routing calculation, a stable relay backbone network and a data transmission path are constructed, the packet loss rate during network transmission is reduced, the network throughput is improved, and the application has a certain application value in the flight ad hoc network composed of aerial unmanned aerial vehicle nodes.

Description

Technical Field

[0001] This invention relates to the field of flight ad hoc network communication technology, and specifically to a method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness. Background Technology

[0002] Flying Ad Hoc Networks (FANETs) typically consist of multiple flying nodes that exchange information and collaborate to complete tasks via air-to-air links. As UAVs and unmanned system platforms evolve towards higher speeds and greater mobility, FANETs have significant application value in scenarios such as military reconnaissance, emergency communication, disaster monitoring, and UAV swarm collaboration. However, this also places higher demands on the stability and reliability of network communication.

[0003] Compared to traditional Mobile Ad Hoc Networks (MANETs), nodes in flight ad hoc networks move faster and are more mobile, resulting in more frequent network topology changes. Under conditions of limited node communication radius, complex mission airspace, and drastic relative node movement, network link states are prone to rapid fluctuations, leading to frequent link establishment and failure. This, in turn, causes problems such as frequent route reconfiguration, increased packet loss rate, and increased end-to-end latency, affecting the stable transmission of service data.

[0004] Optimized Link State Routing (OLSR) is a typical proactive routing protocol that discovers and maintains network topology by periodically exchanging HELLO and TC messages and employs a multipoint relay (MPR) mechanism to reduce control message forwarding overhead. However, existing OLSR protocols still suffer from insufficient adaptability when directly applied to highly dynamic ad hoc networks. On the one hand, traditional MPR selection mechanisms primarily rely on neighbor coverage capabilities, failing to adequately consider link robustness and sustained stability, potentially selecting relay links that are about to deteriorate or disconnect. On the other hand, routing methods based on minimum hop count struggle to guarantee long lifespan and high transmission reliability for selected paths in highly dynamic scenarios.

[0005] Therefore, existing ad hoc network routing protocols are insufficient to meet the needs of highly dynamic flying ad hoc networks in terms of rapid topology changes, insufficient link stability, and reliable transmission. There is an urgent need to propose a routing protocol suitable for highly dynamic flying ad hoc networks to improve the stability of link selection and path construction, reduce the probability of route failure, and ensure the reliability of data transmission. Summary of the Invention

[0006] To address the issues of frequent link disconnections and transmission instability in highly dynamic flight ad hoc networks (QAFs), this invention proposes an improved OLSR routing protocol for QAFs (CLR-OLSR) that is aware of link robustness. This protocol normalizes and weights fused Received Signal Strength Indicator (RSSI) and Link Lifetime (LLifetime) to construct a composite link robustness index. This index is then used to optimize MPR selection and route calculation, building a stable relay backbone network and data transmission paths, reducing packet loss during network transmission, and improving network throughput.

[0007] The technical solution of the present invention is as follows: a method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness, applied to a flight ad hoc network composed of aerial drone nodes, to construct a stable relay backbone network and data transmission path; The flight ad hoc network is composed of The system consists of several UAV nodes. The reachability of the links between UAV nodes is determined by the physical layer reception conditions. The node motion model adopts the Gauss-Markov motion model. The motion of each UAV node is independent and does not interfere with each other. In a short time scale, the speed of the UAV node does not change abruptly, and its high-speed motion trajectory is uniform linear motion. Each drone node uses an omnidirectional antenna with the same transmission power; The MAC layer adopts the IEEE 802.11 protocol. After the routing decision is completed, each node allocates channels based on the MAC layer protocol. The steps include the following: Step 1: Discover neighbor nodes and obtain their location information, speed information, and received signal strength. Calculate the link lifetime based on the location and speed information of the neighbor nodes. Perform normalized weighted fusion based on the received signal strength and link lifetime to calculate the link robustness index. Step 2: Based on the extended HELLO message and extended TC message, update the local neighbor set and topology set, and write the calculated link robustness index into the extended HELLO message and extended TC message, so that the node can be aware of the link robustness index. Step 3: Based on the link robustness index, improve the MPR selection and route calculation process, and select robust relay backbone nodes and transmission paths that meet the link robustness requirements.

[0008] The specific calculation method for the received signal strength in step one is as follows: The received signal strength is the instantaneous signal power received by the node, representing the instantaneous robustness of the link; for high-altitude ad hoc networks, the channel model is considered as a free-space path loss model; distance from the transmitter. The received signal strength at that location is written as: (1) in, To receive signal strength, For transmission power, For the transmit antenna gain, For receiving antenna gain, For the signal wavelength, by Give, At the speed of light, For signal frequency, The distance between the two drone nodes. This refers to system losses.

[0009] The specific calculation method for the link lifetime in step one is as follows: Link lifetime is defined as the time interval during which two adjacent nodes remain within communication range, characterizing the change in link robustness over time; for a pair of UAV nodes... and drone nodes ,exist The coordinates of the time are respectively and The velocity vector is and ; drone nodes Considered relative to drone nodes move, The relative position at any given moment is drone nodes With drone nodes The distance between them is calculated as follows: (2) The relative speed is: (3) in, and Each is a drone node The magnitude of the velocity vector and the UAV node The magnitude of the velocity vector and For drone nodes The angle between the direction of motion and the horizontal direction and the UAV node The angle between the direction of motion and the horizontal direction; Over time Then, the relative position is calculated as follows: (4) (5) in, Indicates drone node and drone nodes The angle between the directions of motion; (6) When the distance between drone nodes exceeds the communication range, the link is considered broken; from until the link breaks time interval That is, the link lifetime; the boundary conditions are: (7) Substituting equations (4) and (5) into equation (7) yields: (8) The solution to this equation is the link lifetime, expressed as: (9) in, , , This refers to the effective communication range of the drone node.

[0010] The specific calculation method for the link robustness index in step one is as follows: The robustness of the composite link is obtained by normalizing and weighting the received signal strength and link lifetime; the normalization of the received signal strength is as follows: (10) in, This refers to the relative received signal strength. The set received signal strength threshold, This represents the maximum received signal strength in the network. The normalization of link lifetime is as follows: (11) in, For relative link lifetime, The lifetime of the link after truncation processing. The maximum link lifetime is set. For those with For a drone node with several optional links, the link with higher robustness is selected as its next hop, as shown below: (12) Among them, weight and ,and ; and Indicates the number of drone nodes The normalized relative received signal strength of the selected links and the UAV node's first... Normalized relative link lifetime of each optional link. This indicates the robustness of the link.

[0011] The extended HELLO message format and extended TC message format described in step two are as follows: The extended HELLO message includes: reserved fields, sending interval, node forwarding intention link status code, link status information size, neighbor node interface address, newly added node location information, newly added node speed information, and newly added link robustness. The newly added node location information and newly added node speed information are used to transmit the location and speed information of the sending node to its neighbor nodes. After receiving the extended HELLO message, the neighbor node calculates the link lifetime based on its own information. The newly added link robustness indicates the link robustness between the sending node and the receiving node. The extended TC message includes: broadcast neighbor sequence number, reserved fields, advertise neighbor primary address field, and added link robustness, indicating the link robustness between the MPR node and the neighbor node corresponding to the advertise neighbor primary address field.

[0012] The improved MPR selection process described in step three is as follows: Improving MPR selection by combining link robustness considers both the link robustness of nodes and the number of covered nodes, avoiding the selection of nodes with many covered nodes but poor link robustness as candidate MPR nodes. Specifically, this includes the following steps: Step S1: Establish drone node MPR node set and initialize it to an empty set; obtain drone nodes. One-jump neighbor set and two-hop neighbor set ; Step S2: For Perform initial filtering; removal is only possible based on the desired outcome. One-hop neighbor reached drone node, drone node themselves and drone nodes All symmetric one-hop neighbors; Step S3: Traverse Among the candidate MPR nodes, if the forwarding intention value of a certain MPR node is... Then add the MPR node. In, and from Remove the two-hop neighbor nodes that it can cover; Step S4: Traverse Determine if there exists a two-hop node that can only be reached through a certain one-hop neighbor. If so, add the corresponding one-hop neighbor to the list. and from Remove the two-hop neighbor nodes that it can cover; Step S5: Determine the current situation Is it empty? If it is empty, output... Otherwise, proceed to step S6; Step S6: Iterate through all those belonging to But not yet joined The candidate nodes are identified, and their contribution values ​​are calculated according to the contribution calculation formula. ; Step S7: Compare the contribution values ​​of each candidate node. Select the node with the largest contribution value as the current candidate MPR node, and update the maximum contribution value. and its corresponding node forwarding willingness value; Step S8: If the contribution values ​​of multiple candidate nodes differ by less than the threshold, then further compare the forwarding intention values ​​of these nodes and select the node with the better forwarding intention value as the final candidate MPR node. Step S9: Add the finally selected candidate MPR nodes. and from Remove its newly covered two-hop neighbor node, then return to step S5, until... Empty.

[0013] The contribution value of the MPR node in step S6 Defined as: (13) in, For drone nodes The number of currently uncovered two-hop neighbor nodes covered. For the current node and the node Inter-link robustness, It is a constant used for adjustment The weight.

[0014] The improved route calculation process described in step three is as follows: The routing calculation process is improved by incorporating link stability. This involves optimizing route selection by considering both link stability and hop count to choose the shortest route capable of stable transmission. Specifically, this includes the following steps: Step R1: Initialize the node Clear all existing routing table entries. Step R2: Obtain nodes One-jump neighbor set Two-hop neighbor set and topology information set ; Step R3: Traverse Determine each one-hop neighbor node Whether it is a symmetric neighbor and whether its link state is valid; if the node If the conditions are met, add the corresponding one-hop routing table entry to the routing table; Step R4: Traverse According to formula (12), the distance to each two-hop neighbor node is calculated. routing stability ;like If the traversal is complete, proceed to step R8; otherwise, proceed to step R5. Step R5: Determine the two-hop neighbor nodes Is it simultaneously a one-hop neighbor node? If the neighbor is the symmetric neighbor, proceed to step R6; otherwise, proceed to step R7. Step R6: Based on routing quality thresholds Update the existing routes in the routing table and proceed to step R4; Step R7: If node If the route does not already exist in the routing table, or if the new route is of better quality, add or update the two-hop route in the routing table and proceed to step R4. Step R8: Traverse the topology information set Each destination node in Determine if the route to the destination node already exists in the routing table. The route to the previous hop node; if the condition is met, proceed to step R9; otherwise, proceed to step R10. Step R9: Calculate the current candidate multi-hop routes according to the routing quality calculation formula. and at the destination node If the route does not already exist in the routing table or if a new route is better, add or update it to the routing table, then proceed to step R8. Step R10: No new routes are added or updated, output node The routing table.

[0015] The routing table described in step R3 consists of five parts: the destination node address, the next-hop address, the interface index of the destination node, and... Hop count and minimum link robustness of the added entire path For an entire path, the robustness of the links along that path is determined by the hop with the lowest robustness, defined as: (14) in, For the current route Link robustness of hop links; when A value of 0 indicates that the route is invalid and will be removed from the routing table.

[0016] The beneficial effects of this invention are as follows: This invention proposes a high-dynamic flight ad hoc network routing protocol construction method (CLR-OLSR) oriented towards link robustness. It constructs a composite link robustness index based on received signal strength and link lifetime, and extends the HELLO and TC message formats. By combining link robustness optimization with MPR selection and route calculation, a stable relay backbone network and data transmission path are constructed. Compared with the three traditional ad hoc network routing protocols AODV, DSDV, and OLSR, this invention can effectively reduce packet loss rate during network transmission and improve network throughput, exhibiting stronger effectiveness and reliability in highly dynamic environments. Attached Figure Description

[0017] Figure 1 This is a system block diagram of the present invention.

[0018] Figure 2 This is a network model structure diagram of the present invention.

[0019] Figure 3 This is a diagram of the expanded HELLO message format.

[0020] Figure 4 This is a diagram of the expanded TC message format.

[0021] Figure 5 This invention compares the packet loss rate with three routing protocols—AODV, DSDV, and OLSR—under varying speed conditions.

[0022] Figure 6 This invention compares the throughput of the present invention with that of three routing protocols, AODV, DSDV, and OLSR, under varying speed conditions.

[0023] Figure 7 This invention compares the packet loss rate with three routing protocols—AODV, DSDV, and OLSR—under varying node numbers.

[0024] Figure 8 This invention compares the throughput of the present invention with that of three routing protocols, AODV, DSDV, and OLSR, under varying node numbers. Detailed Implementation

[0025] A method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness is applied to a flight ad hoc network composed of aerial UAV nodes to build a stable relay backbone network and data transmission path. The flight ad hoc network is composed of The system consists of several UAV nodes, each performing tasks and communicating with each other within a pre-defined task area. The reachability of links between UAV nodes is determined by the physical layer reception conditions. When a node moves outside the effective communication range, the corresponding link is considered broken. The node motion model adopts the Gauss-Markov Mobility Model. The motion of each UAV node is independent and does not interfere with each other. Within a short time scale, the speed of the UAV node does not change abruptly, and its high-speed trajectory is uniform linear motion. Each drone node uses an omnidirectional antenna with the same transmission power; The MAC layer adopts the IEEE 802.11 protocol. After the routing decision is completed, each node allocates channels based on the MAC layer protocol. By using the CLR-OLSR protocol described in this invention, combined with the proposed link robustness optimization MPR selection and routing calculation process, the transmission packet loss rate can be reduced and the network throughput can be improved.

[0026] Specifically, the following steps are included: Step 1: Discover neighbor nodes and obtain their location information, speed information, and received signal strength. Calculate the link lifetime based on the location and speed information. Perform normalized weighted fusion based on the received signal strength and link lifetime to calculate the link robustness index. Define link lifetime to characterize the link's ability to maintain robustness over a future period. Step 2: Based on the extended HELLO message and extended TC message, update the local neighbor set and topology set, and write the calculated link robustness index into the extended HELLO message and extended TC message, so that the node can be aware of the link robustness index. Step 3: Based on the link robustness index, improve the MPR selection and route calculation process, and select robust relay backbone nodes and transmission paths that meet the link robustness requirements.

[0027] The specific calculation method for the received signal strength in step one is as follows: The received signal strength is the instantaneous signal power received by the node, representing the instantaneous robustness of the link; for high-altitude ad hoc networks, the channel model is considered as a free-space path loss model; distance from the transmitter. The received signal strength at that location is written as: in, To receive signal strength, For transmission power, For the transmit antenna gain, For receiving antenna gain, For the signal wavelength, by Give, At the speed of light, For signal frequency, The distance between the two drone nodes. The system loss is assumed to be 1 in this invention.

[0028] When a drone node receives a HELLO message from its neighboring node, the node will obtain the received signal strength from its physical layer and compare the measured value with a threshold. The link will be preserved only if the received signal strength meets the following conditions, based on the comparison.

[0029] In step one, the link lifetime is defined as the time interval during which two adjacent nodes remain within communication range, characterizing the change in link robustness over time. For high-speed moving drones, their speed does not change suddenly, and their trajectory over a short period can be approximated as uniform linear motion. This invention assumes that the altitude difference between drones is related to their position within the communication range. The horizontal distance on the plane is negligible compared to that between a pair of nodes. and nodes They are in The coordinates of the time are respectively and The velocity vector is and ; drone nodes Considered relative to drone nodes move, The relative position at any given moment is drone nodes With drone nodes The distance between them is calculated as follows: The relative speed is: in, and Each is a drone node The magnitude of the velocity vector and the UAV node The magnitude of the velocity vector and For drone nodes The angle between the direction of motion and the horizontal direction and the UAV node The angle between the direction of motion and the horizontal direction; Over time Then, the relative position is calculated as follows: in, Indicates drone node and drone nodes The angle between the directions of motion; When the distance between drone nodes exceeds the communication range, the link is considered broken; from until the link breaks time interval That is, the link lifetime; the boundary conditions are: The solution to this equation is the link lifetime, expressed as: in, , , This refers to the effective communication range of the drone node.

[0030] Furthermore, by normalizing and weighting the received signal strength and link lifetime, the robustness of the composite link can be obtained. The normalization process for the received signal strength is as follows: in, This refers to the relative received signal strength. The set received signal strength threshold, This represents the maximum received signal strength in the network.

[0031] The normalization of link lifetime is as follows: in, For relative link lifetime, The lifetime of the link after truncation processing. This is the maximum link lifetime set.

[0032] Furthermore, after normalization, the robustness of the weighted composite link can be obtained. For links with... For a drone node with an optional link, the link with higher robustness is selected as its next hop, denoted as... Among them, weight and ,and . and Indicates the node number Normalized relative received signal strength and relative link lifetime for each selectable link. This indicates the link's robustness. The constructed composite link robustness will be used to optimize MPR selection and route calculation.

[0033] In the CLR-OLSR protocol described in this invention, nodes exchange neighbor information and link status information through HELLO messages and TC messages, so it is necessary to extend the format of HELLO messages and TC messages.

[0034] The extended HELLO message includes the following fields: Reserved, Htime (sending interval), Willingness (node ​​forwarding intention), Link Code, LinkMessage Size, Neighbor Interface Address, and newly added Node Position and Node Velocity. These are used to transmit the sender node's position and velocity information to its neighbors, allowing neighboring nodes to calculate the link lifetime based on their own information. The newly added Link Robustness indicates the robustness of the link between the sender and receiver nodes and can be calculated using the previously described scheme. Received signal strength can be directly read from the physical layer and therefore does not need to be added as a new field to the HELLO message.

[0035] TC messages are used to transmit network topology information. Nodes construct the entire network topology locally based on the received TC messages. The extended TC message includes the following fields: ANSW broadcast neighbor sequence number, Reserved field, Advertised Neighbor Main Address field, and the newly added Link Robustness field, which represents the link robustness between the MPR node and the neighbor node corresponding to the Advertised Neighbor Main Address, calculated by the scheme described above.

[0036] Furthermore, the control messages of the CLR-OLSR protocol described in this invention are periodically transmitted in the network. When a node receives a HELLO message from a neighboring node, it adds or updates the neighboring node's information to its local one-hop neighbor set. Based on the location and speed fields carried in the HELLO message, it calculates the link lifetime and sets the effective time of the link with that node as the calculated value, with a value range of [range missing]. The node reads the received signal strength from the physical layer, calculates the link robustness, and populates it into the one-hop neighbor set. Then, it parses the link information in the HELLO message, reads the address and link robustness, and updates the corresponding two-hop neighbor set. TC messages are only forwarded by MPR nodes. When a node receives a TC message, it reads the address information and link robustness and updates the corresponding local topology set.

[0037] The MPR selection method, which incorporates improved link robustness, considers both the link robustness of a node and the number of covered nodes. It avoids selecting nodes with many covered nodes but poor link robustness as candidate MPR nodes. Specifically, it includes the following steps: Step S1: Create a node MPR node set and initialize it to an empty set; get the node One-jump neighbor set and two-hop neighbor set .

[0038] Step S2: Perform initial filtering, removing items that can only be selected based on a preference value. Nodes reached by one-hop neighbors, nodes itself and nodes All symmetrical one-hop neighbors.

[0039] Step S3: Traverse Among the candidate MPR nodes, if a node's forwarding intention value is Then add the node. In, and from Remove the two-hop neighbor nodes that it can cover.

[0040] Step S4: Traverse Determine if there exists a two-hop node that can only be reached through a certain one-hop neighbor. If so, add the corresponding one-hop neighbor to the list. and from Remove the two-hop neighbor nodes that it can cover.

[0041] Step S5: Determine the current situation Check if the value is empty. If it is empty, the algorithm ends and outputs the result. Otherwise, continue with step S6.

[0042] Step S6: Iterate through all those belonging to But not yet joined The candidate nodes are identified, and their contribution values ​​are calculated according to the contribution calculation formula. .

[0043] Step S7: Compare the contribution values ​​of each candidate node. Select the node with the largest contribution value as the current candidate MPR node, and update the maximum contribution value. And its corresponding node forwarding willingness value.

[0044] Step S8: If the contribution values ​​of multiple candidate nodes differ by less than the threshold, then further compare the forwarding intention values ​​of these nodes and select the node with the better forwarding intention value as the final candidate MPR node.

[0045] Step S9: Add the finally selected candidate MPR nodes. and from Remove its newly covered two-hop neighbor node, then return to step S5, until... Empty.

[0046] Specifically, the contribution of the MPR node in step S6 Defined as in, For nodes The number of currently uncovered two-hop neighbor nodes covered. For the current node and the node Inter-link robustness, It is a constant used for adjustment The weight.

[0047] The improved routing calculation method, which incorporates link stability improvements, simultaneously considers link stability and hop count to optimize route selection and choose the shortest route capable of stable transmission. Specifically, it includes the following steps: Step R1: Initialize the node Clear the routing table of all existing routing entries.

[0048] Step R2: Obtain nodes One-jump neighbor set Two-hop neighbor set and topology information set .

[0049] Step R3: Traverse Determine each one-hop neighbor node Whether it is a symmetric neighbor and whether its link state is valid; if the node If the conditions are met, a corresponding one-hop routing table entry is added to the routing table.

[0050] Step R4: Traverse The routing quality is calculated based on the formula for each two-hop neighbor node. routing stability ;like If the traversal is complete, proceed to step R8; otherwise, proceed to step R5.

[0051] Step R5: Determine the two-hop neighbor nodes Is it simultaneously a one-hop neighbor node? If the neighbor is the symmetric neighbor, proceed to step R6; otherwise, proceed to step R7.

[0052] Step R6: Based on routing quality thresholds Update the existing routes in the routing table and proceed to step R4.

[0053] Step R7: If node If the route does not already exist in the routing table, or if the new route is of better quality, add or update the two-hop route to the routing table and proceed to step R4.

[0054] Step R8: Traverse the topology information set Each destination node in Determine if the destination node already exists in the routing table. The route to the previous hop node; if the condition is met, proceed to step R9; otherwise, proceed to step R10.

[0055] Step R9: Calculate the current candidate multi-hop routes according to the routing quality calculation formula. and at the node If the route does not already exist in the routing table or if a new route is better, add or update it to the routing table and proceed to step R8.

[0056] Step R10: No new routes are added or updated, output node The routing table.

[0057] Specifically, the routing table described in step R3 consists of five parts: Destination Address, NextHop Address, Interface Index of the Destination Node, and so on. Hop count and minimum link robustness of the added entire path For a given path, the robustness of all links along that path is determined by the hop with the lowest robustness, defined as follows: in, For the link robustness of the m-th hop link of the current route. When A value of 0 indicates that the route is invalid and will be removed from the routing table.

[0058] The present invention will be further described below with reference to specific embodiments. The protocol described in this invention is implemented in NS-3.37. In order to evaluate the performance of the protocol in highly dynamic scenarios, the network performance of the protocol under different speeds and different numbers of nodes is simulated and analyzed. The performance of three classic routing protocols, OLSR, AODV and DSDV, is simulated under the same scenario as a benchmark comparison.

[0059] Figure 2 In the network model shown, the drone speed is set. The speed range is 100 m / s to 700 m / s, with a velocity interval of 100 m / s. The nodes are initially randomly distributed at a distance of 10000 m. Within a 10,000 m area, the initial velocity magnitude is at the current simulated velocity. Random distribution within the neighborhood of, i.e. m / s, random direction. Transmission range The distance is 4000 m, and the mobility model is a Gaussian Markov mobility model. The MAC layer protocol is IEEE 802.11. The total simulation time is... The transmission time is 140 seconds, with the first 20 seconds used for route establishment. Data transmission begins within the network from the 20th second onwards, using Constant Bit Rate Service Flow (CBR). The transmission rate is 2048 bps, and the data packet size is 96 bytes.

[0060] Figure 5 The paper demonstrates how packet loss rates for each protocol change with speed. As node speed increases, the overall packet loss rate of all four protocols rises. This is because rapid network topology changes and frequent link breaks in highly dynamic scenarios lead to expired routing tables and significant packet loss. DSDV exhibits the highest packet loss rate with a significant increase, indicating that its routes struggle to converge in a timely manner under high dynamic conditions, resulting in numerous outdated routes in its routing table and persistent forwarding failures and packet loss. AODV's packet loss rate is slightly lower than DSDV's, but its on-demand discovery routing struggles to keep up with topology changes in highly dynamic scenarios, causing packet loss. OLSR, which selects the shortest path based solely on hop count, sees its link breakage risk increase with rising speeds. In contrast, the protocol described in this invention maintains the lowest packet loss rate across the entire speed range with the slowest increase, remaining below 10% even at 700 m / s. This is because it incorporates received signal strength and link lifetime constraints, achieving optimal link robustness and longest lifetime through improved MPR selection and routing, effectively reducing the risk of link breakage, significantly improving data delivery reliability, and exhibiting minimal overall packet loss rate variation.

[0061] Figure 6The diagram illustrates the throughput variations of each protocol with speed. As node speed increases, the throughput of all three traditional protocols shows a significant downward trend, reflecting the increased frequency of link failures under high dynamic conditions, resulting in substantial packet loss and retransmissions, ultimately leading to a deterioration in overall network throughput. The protocol described in this invention maintains the highest and most stable throughput across all speeds. This is because this invention introduces a link robustness-aware mechanism, avoiding unstable links in MPR selection and routing, prioritizing the construction of more reliable forwarding backbones, reducing forwarding failures and MAC layer retransmissions, ensuring the quality of path data transmission, and effectively mitigating the negative impacts of high dynamics.

[0062] Figures 7 and 8 show the packet loss rate and network throughput performance of each protocol under different numbers of nodes. DSDV and AODV, limited by their own topology maintenance and route discovery mechanisms, exhibit high packet loss rates and low throughput. OLSR, relying solely on hop count for route selection, still suffers from significant packet loss, and its throughput performance is only at a moderate level. The protocol proposed in this invention prioritizes link stability, avoiding the selection of links with low stability. By selecting stable routes, it minimizes packet loss, consistently keeping the packet loss rate below 10%, while maintaining high throughput during simulation, demonstrating a clear advantage.

[0063] The simulation results above demonstrate that the high-dynamic flight ad hoc network routing protocol of this invention, which focuses on link robustness, constructs a composite link robustness index by fusing received signal strength and link lifetime, and improves the MPR selection and routing calculation processes based on this index. The protocol outperforms comparative protocols in terms of packet loss rate and throughput under different node speeds and densities. Therefore, this invention can effectively improve link stability and data transmission reliability in high-dynamic flight ad hoc network scenarios, and is suitable for flight ad hoc network communication scenarios composed of high-speed moving UAV nodes.

[0064] The above-described embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. A method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness, applied to a flight ad hoc network composed of aerial unmanned aerial vehicle (UAV) nodes, characterized in that, Construct a stable relay backbone network and data transmission paths; The flight ad hoc network is composed of The system consists of several UAV nodes. The reachability of the links between UAV nodes is determined by the physical layer reception conditions. The node motion model adopts the Gauss-Markov motion model. The motion of each UAV node is independent and does not interfere with each other. In a short time scale, the speed of the UAV node does not change abruptly, and its high-speed motion trajectory is uniform linear motion. Each drone node uses an omnidirectional antenna with the same transmission power; The MAC layer adopts the IEEE 802.11 protocol. After the routing decision is completed, each node allocates channels based on the MAC layer protocol. The steps include the following: Step 1: Discover neighbor nodes and obtain their location information, speed information, and received signal strength. Calculate the link lifetime based on the location and speed information of the neighbor nodes. Perform normalized weighted fusion based on the received signal strength and link lifetime to calculate the link robustness index. Step 2: Based on the extended HELLO message and extended TC message, update the local neighbor set and topology set, and write the calculated link robustness index into the extended HELLO message and extended TC message, so that the node can be aware of the link robustness index. Step 3: Based on the link robustness index, improve the MPR selection and route calculation process, and select robust relay backbone nodes and transmission paths that meet the link robustness requirements.

2. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The specific calculation method for the received signal strength in step one is as follows: The received signal strength is the instantaneous signal power received by the node, representing the instantaneous robustness of the link; for high-altitude ad hoc networks, the channel model is considered as a free-space path loss model; distance from the transmitter. The received signal strength at that location is written as: （1） in, To receive signal strength, For transmission power, For the transmit antenna gain, For receiving antenna gain, For the signal wavelength, by Give, At the speed of light, For signal frequency, The distance between the two drone nodes. This refers to system losses.

3. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The specific calculation method for the link lifetime in step one is as follows: Link lifetime is defined as the time interval during which two adjacent nodes remain within communication range, characterizing the change in link robustness over time; for a pair of UAV nodes... and drone nodes ,exist The coordinates of the time are respectively and The velocity vector is and ; drone nodes Considered relative to drone nodes move, The relative position at any given moment is drone nodes With drone nodes The distance between them is calculated as follows: （2） The relative speed is: （3） in, and Each is a drone node The magnitude of the velocity vector and the UAV node The magnitude of the velocity vector and For drone nodes The angle between the direction of motion and the horizontal direction and the UAV node The angle between the direction of motion and the horizontal direction; Over time Then, the relative position is calculated as follows: （4） （5） in, Indicates drone node and drone nodes The angle between the directions of motion; （6） When the distance between drone nodes exceeds the communication range, the link is considered broken; from until the link breaks time interval That is, the link lifetime; the boundary conditions are: （7） Substituting equations (4) and (5) into equation (7) yields: （8） The solution to this equation is the link lifetime, expressed as: （9） in, , , This refers to the effective communication range of the drone node.

4. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The specific calculation method for the link robustness index in step one is as follows: The robustness of the composite link is obtained by normalizing and weighting the received signal strength and link lifetime; the normalization of the received signal strength is as follows: （10） in, This refers to the relative received signal strength. The set received signal strength threshold, This represents the maximum received signal strength in the network. The normalization of link lifetime is as follows: （11） in, For relative link lifetime, The lifetime of the link after truncation processing. The maximum link lifetime is set. For those with For a drone node with several optional links, the link with higher robustness is selected as its next hop, as shown below: （12） Among them, weight and ,and ; and Indicates the number of drone nodes The normalized relative received signal strength of the selected links and the UAV node's first... Normalized relative link lifetime of each optional link. This indicates the robustness of the link.

5. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The extended HELLO message format and extended TC message format described in step two are as follows: The expanded HELLO message includes: reserved fields, sending interval, node forwarding intention link status code, link status information size, neighbor node interface address, newly added node location information, newly added node speed information, and newly added link robustness; the newly added node location information and newly added node speed information are used to transmit the location and speed information of the sending node to its neighbor nodes; After receiving the extended HELLO message, neighboring nodes calculate the link lifetime based on their own information; a new link robustness is added, representing the magnitude of link robustness between the sending node and the receiving node; The extended TC message includes: broadcast neighbor sequence number, reserved fields, advertise neighbor primary address field, and added link robustness, indicating the link robustness between the MPR node and the neighbor node corresponding to the advertise neighbor primary address field.

6. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The improved MPR selection process described in step three is as follows: Improving MPR selection by combining link robustness considers both the link robustness of nodes and the number of covered nodes, avoiding the selection of nodes with many covered nodes but poor link robustness as candidate MPR nodes. Specifically, this includes the following steps: Step S1: Establish drone node MPR node set and initialize it to an empty set; obtain drone nodes. One-jump neighbor set and two-hop neighbor set ; Step S2: For Perform initial filtering; removal is only possible based on the desired outcome. One-hop neighbor reached drone node, drone node themselves and drone nodes All symmetric one-hop neighbors; Step S3: Traverse Among the candidate MPR nodes, if the forwarding intention value of a certain MPR node is... Then add the MPR node. In, and from Remove the two-hop neighbor nodes that it can cover; Step S4: Traverse Determine if there exists a two-hop node that can only be reached through a certain one-hop neighbor. If so, add the corresponding one-hop neighbor to the list. and from Remove the two-hop neighbor nodes that it can cover; Step S5: Determine the current situation Is it empty? If it is empty, output... Otherwise, proceed to step S6; Step S6: Iterate through all those belonging to But not yet joined The candidate nodes are identified, and their contribution values ​​are calculated according to the contribution calculation formula. ; Step S7: Compare the contribution values ​​of each candidate node. Select the node with the largest contribution value as the current candidate MPR node, and update the maximum contribution value. and its corresponding node forwarding willingness value; Step S8: If the contribution values ​​of multiple candidate nodes differ by less than the threshold, then further compare the forwarding intention values ​​of these nodes and select the node with the better forwarding intention value as the final candidate MPR node. Step S9: Add the finally selected candidate MPR nodes. and from Remove its newly covered two-hop neighbor node, then return to step S5, until... Empty.

7. The method for constructing a highly dynamic flight ad hoc network routing protocol for link robustness according to claim 6, wherein the contribution value of the MPR node in step S6 is... Defined as: （13） in, For drone nodes The number of currently uncovered two-hop neighbor nodes covered. For the current node and the node Inter-link robustness, It is a constant used for adjustment The weight.

8. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 1, characterized in that, The improved route calculation process described in step three is as follows: The routing calculation process is improved by incorporating link stability. This involves optimizing route selection by considering both link stability and hop count to choose the shortest route capable of stable transmission. Specifically, this includes the following steps: Step R1: Initialize the node Clear all existing routing table entries. Step R2: Obtain nodes One-jump neighbor set Two-hop neighbor set and topology information set ; Step R3: Traverse Determine each one-hop neighbor node Whether it is a symmetric neighbor and whether its link state is valid; if the node If the conditions are met, add the corresponding one-hop routing table entry to the routing table; Step R4: Traverse According to formula (12), the distance to each two-hop neighbor node is calculated. routing stability ;like If the traversal is complete, proceed to step R8; otherwise, proceed to step R5. Step R5: Determine the two-hop neighbor nodes Is it simultaneously a one-hop neighbor node? If the neighbor is the symmetric neighbor, proceed to step R6; otherwise, proceed to step R7. Step R6: Based on routing quality thresholds Update the existing routes in the routing table and proceed to step R4; Step R7: If node If the route does not already exist in the routing table, or if the new route is of better quality, add or update the two-hop route in the routing table and proceed to step R4. Step R8: Traverse the topology information set Each destination node in Determine if the route to the destination node already exists in the routing table. The route to the previous hop node; if the condition is met, proceed to step R9; otherwise, proceed to step R10. Step R9: Calculate the current candidate multi-hop routes according to the routing quality calculation formula. and at the destination node If the route does not already exist in the routing table or if a new route is better, add or update it to the routing table, then proceed to step R8. Step R10: No new routes are added or updated, output node The routing table.

9. The method for constructing a highly dynamic flight ad hoc network routing protocol oriented towards link robustness according to claim 8, characterized in that, The routing table described in step R3 consists of five parts: the destination node address, the next-hop address, the interface index of the destination node, and... Hop count and minimum link robustness of the added entire path ; For an entire path, the robustness of the links along that path is determined by the hop with the lowest robustness, defined as: （14） in, For the current route number Link robustness of hop links; when A value of 0 indicates that the route is invalid and will be removed from the routing table.

Images