A multipath routing protocol method for mobility prediction in unmanned aerial vehicle ad hoc networks
By establishing a master backup route in the drone ad hoc network and using the Gaussmarkov mobile model to predict the node location and link failure time, the problem of frequent route breakage in the drone ad hoc network is solved, and the stability of the network is improved and transmission rate.
Patent Information
- Application Number
- CN202210137735.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-15
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2042-02-15
AI Technical Summary
The existing drone ad hoc network routing protocols are difficult to cope with the frequent routing breakage caused by high-speed movement and rapid changes in topology, resulting in low network transmission rates and inability to find backup paths in time.
A multipath routing protocol for predicting mobility of drone ad hoc networks is proposed. By establishing a master backup route, using the Gaussmarkov mobile model to predict node location and link failure time, switching to backup routes in time, reducing routing reconstruction time.
It effectively reduces the routing reconstruction time, improves network reliability and transmission rate, reduces routing overhead, and improves the stability of the UAV ad hoc network in a high-speed movement and fast topological change environment.
Smart Images

Figure CN114390631B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of communication technology, and in particular relates to a multipath routing protocol method for predicting the mobility of an unmanned aerial vehicle self-organizing network. Background Art
[0002] With the development of wireless ad hoc network research and multi-UAV cooperative operations, UAV ad hoc networks have become a new research hotspot. The high-speed mobility of nodes and the high dynamics of network topology in complex scenarios have brought new challenges to the communication networking technology of UAV ad hoc networks. As an informationized combat platform, UAVs play an important role in information collection, intelligence acquisition, and communication relay. UAV clusters can provide a series of functions such as virtual air bases, intelligent platforms, and electronic warfare, giving full play to the advantages of information networks.
[0003] UAV Ad Hoc networks (UANET) are the key technology to achieve collaborative integration. They can share information obtained by all UAVs in the region through wireless channels, improve the UAV network system's ability to process information in real time and respond to special situations, and enhance the survivability of UAVs in different environments. Ensuring real-time and reliable communication between UAVs is the basis for other functions of UAV Ad Hoc networks. How to achieve an efficient communication system has been the focus of research on UAV Ad Hoc networks in recent years.
[0004] Dynamic topology changes in UAV self-organizing networks UAV nodes move in the network at high speed and in an arbitrary manner, and are affected by comprehensive factors such as mutual interference between wireless channels, terrain and weather. The network topology formed by UAVs through wireless channels may change at any time, and the way and speed of change are difficult to predict. To complete complex tasks, UAVs need to form UAV self-organizing networks, that is, multiple wireless mobile communication networks dynamically formed by several UAVs without the help of any central station. Mobile Ad Hoc networks (MANETs) are multi-hop temporary networks formed by mobile nodes. Their centerless and self-organizing characteristics are particularly suitable for the needs of UAV groups working together in different environments, among which routing technology is the core technology. At present, there are many routing technologies used in mobile Ad Hoc networks, but they are not suitable for UAV self-organizing networks with high-speed movement, rapid changes in topology, and continuous joining or leaving of nodes. Therefore, it is of great significance to design an efficient routing protocol suitable for UAV self-organizing networks to achieve real-time connectivity of wireless communication links. However, existing research only considers the single characteristic of drones and does not solve the problem of frequent route disconnection caused by the dynamic changes in the height of drones. Since single-path routing does not make full use of the effective information of the network, it increases a considerable routing overhead. Moreover, data transmission cannot be transmitted in parallel, or a backup path cannot be found in time when the link is disconnected, resulting in a low network transmission rate. Summary of the invention
[0005] In order to solve the problems existing in the above-mentioned prior art, the present invention proposes a multipath routing protocol method for mobility prediction of unmanned aerial vehicle self-organizing network, the method comprising: establishing a unmanned aerial vehicle mobile node system model according to the unmanned aerial vehicle mobile path; using a multipath routing algorithm to establish a primary backup route according to the constructed unmanned aerial vehicle mobile node system model, wherein the primary backup route includes a primary path and a backup path; obtaining information of the node to be predicted, inputting the node information into a node mobility prediction model, and obtaining a mobility prediction result of the node; and performing route switching according to the mobility prediction result.
[0006] Preferably, the process of establishing the primary and backup routes using the multipath routing algorithm includes: establishing multiple available routes between the source node and the destination node, and taking the route with the best link quality as the primary route; using the Gauss-Markov mobility model to predict the location of the node at the next moment based on the residual energy of the node, and switching to the backup route in advance when it is predicted that the neighboring node is out of communication range or the link is about to fail.
[0007] Preferably, the process of inputting the node information into the mobility prediction model includes: the node information includes the node's remaining energy, link failure time, and the node's signal strength;
[0008] The residual energy of the node is input into the node mobility prediction model based on the residual energy to perform mobility prediction; if the predicted stability of the neighboring node is less than zero, a warning value is set for the neighboring node; if the predicted stability of the neighboring node is greater than zero, the neighboring node is used as a switching route;
[0009] The Gauss-Markov mobility model is used to obtain the location and mobility information of two adjacent nodes, and the link failure time between the two nodes is calculated based on the location and mobility information of the adjacent nodes; the link failure time between the two nodes is input into the node mobility prediction model based on the link failure time to perform mobility prediction; when the link failure time between the two adjacent nodes is predicted to be less than or equal to zero, a warning threshold is set for the neighbor node, otherwise the neighbor node is used as a switching route;
[0010] Obtain the received packets received by each node, and detect the power of the sending node based on the received packets; predict the mobility of the node based on the power of the sending node; if the predicted stability of the neighboring node is less than zero, set a warning value for the neighboring node; if the predicted stability of the neighboring node is greater than zero, use the neighboring node as a switching route.
[0011] Furthermore, the process of obtaining the remaining energy of the node includes: setting the initial energy E0 of the node, calculating the energy E consumed by the node to perform a routing operation i , the remaining energy of the node is calculated based on the initial energy and the consumed energy; the formula for the energy consumed by the node to perform a routing operation is:
[0012] E i =P1T1+P2T2=P1L1 / V1+P2L2 / V2
[0013] Among them, P1 represents the power of node i receiving data, T1 represents the duration of receiving data packets, L1 represents the length of receiving data packets, V1 represents the data receiving rate, P2 represents the power of node i sending data, T2 represents the duration of sending data packets, L2 represents the length of sending data packets, and V2 represents the data sending rate.
[0014] Preferably, the process of calculating the link failure time between two nodes includes: using the GPS system to obtain information about the two nodes m and n, the information including the coordinates of the nodes, the average speed of the node movement, the moving direction of the nodes, and the transmission radius r of the two nodes; calculating the link failure time based on the information of the two nodes.
[0015] Furthermore, the formula for calculating link failure time is:
[0016]
[0017] a=v m cosθm -v n cosθ n
[0018] b = x m -x n
[0019] c=v m sinθ m -v n sinθ n
[0020] d=y m -y n
[0021] Where LET represents the link failure time, a represents the average speed difference between node m and node n in the horizontal direction, b represents the position distance between node m and node n in the horizontal direction, c represents the average speed difference between node m and node n in the vertical direction, d represents the position distance between node m and node n in the vertical direction, and v m represents the average speed of node m, v n represents the average speed of node n’s movement, θ m represents the moving direction of node m, θ n represents the moving direction of node n, x m Represents the horizontal coordinate of node m, x n Represents the horizontal coordinate of node n, y m Represents the ordinate of node m, y n Represents the vertical coordinate of node n.
[0022] Preferably, predicting the mobility of the node according to the power of the sending node includes: obtaining the receiving power of node A received by node B at the current moment and the received power received by node A from node B at the last moment According to the obtained and Calculate the relative motion speed V of the two nodes AB , when V AB When ≥0, node A and node B are close to each other or keep the distance unchanged, and the stability of the two nodes is strong; otherwise, node A and node B are far away from each other, and the stability between the two nodes is weak.
[0023] Furthermore, the relative motion speed formula is:
[0024]
[0025] in, Indicates the receiving power of node A to node B at the current moment; Indicates the received power that node A received from node B at the last moment.
[0026] Preferably, the node executing route switching includes the node periodically sending HELLO messages and TC packets containing predicted location information to detect neighbor nodes in the route list, and when the detected route warning value of the neighbor node is 1, the backup route is enabled, otherwise the backup route is not enabled.
[0027] Advantages of the present invention:
[0028] The multipath routing protocol scheme for mobility prediction of UAV ad hoc networks proposed in the present invention not only takes into account the high dynamics of UAV nodes, the residual energy (Dump Energy, DE) of UAV nodes, signal intensity (Signal Intensity, SI) and link expiration time (Link Expiration Time, LET) but also takes into account the problem of route reconstruction caused by frequent route breaks. Establishing a backup routing mechanism can reduce the time of route reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 This is a flow chart of the multipath routing algorithm for predicting mobility of unmanned aerial vehicle ad hoc networks of the present invention;
[0030] Figure 2 A schematic diagram of link failure time calculation of the present invention;
[0031] Figure 3 This is a comparison diagram of the original OLSR routing table structure and the improved routing table structure;
[0032] Figure 4 It is the motion trajectory diagram of the drone node in ns-3 of the present invention;
[0033] Figure 5 A flowchart of node mobility prediction of the present invention;
[0034] Figure 6 A comparison diagram of the average packet transmission rate between the present invention and the original OLSR;
[0035] Figure 7 This is a comparison chart of the average end-to-end delay between the present invention and the original OLSR;
[0036] Figure 8 This is a comparison diagram of the end-to-end jitter of the present invention and the original OLSR;
[0037] Fig. 9 This is a comparison diagram of the routing overhead of the present invention and the original OLSR. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0039] A multipath routing protocol method for mobility prediction in UAV ad hoc networks, such as Figure 1 The method comprises: establishing a drone mobile node system model according to the drone mobile path; establishing a primary backup route using a multipath routing algorithm according to the constructed drone mobile node system model, wherein the primary backup route includes a primary path and a backup path; obtaining information of a node to be predicted, inputting the node information into a node mobility prediction model, obtaining a mobility prediction result of the node, and constructing a route list according to the prediction result; and executing route switching according to the node in the route list.
[0040] Multipath routing protocols can generally be divided into intersecting multipaths and non-intersecting multipaths according to the intersection of links. Among them, non-intersecting multipaths can be further divided into link non-intersecting multipaths and node non-intersecting multipaths. At the same time, the established multiple paths also have two usage modes, namely the backup path mode and the parallel transmission mode. The present invention proposes a backup routing protocol algorithm, which adopts the main backup path mechanism. An embedded Dijkstra algorithm is adopted on the basis of the original OLSR protocol to directly calculate n non-intersecting paths between the source and the destination node. In this protocol, n=2, and the one with the smallest number of hops is selected as the main path, and the other is used as the backup path. Then, according to the results of the mobile prediction, when the main path is about to fail, it automatically switches to the backup path to reduce the link failure time and increase network reliability.
[0041] According to the constructed UAV mobile node system model, the multipath routing algorithm is used to establish the primary and backup routes, including:
[0042] Step 1: Dijkstra algorithm solves the shortest path problem between any two vertices in a directed graph (or undirected graph). The main feature of the algorithm is that it starts from the starting point and expands outward until it reaches the end point. Dijkstra algorithm is a very representative shortest path algorithm.
[0043] When establishing a backup route, the embedded Dijkstra algorithm is used, that is, the Dijkstra algorithm is used to calculate multiple non-intersecting paths based on the Dijkstra algorithm used when calculating the routing table in the original routing protocol. In addition, the cost in the network topology diagram is all 1, that is, one distance.
[0044] The routing table structure of the original OLSR protocol and the improved routing protocol is as follows Figure 3 The improved routing table structure is as follows: R_dest_addr: destination node address; R_nest_addr: next hop node address; R_dist: number of hops required to pass through the destination node; R_iface_addr: local interface address of the next hop node to reach the destination node; R_backup: next hop node address of the backup path; R_backup_dist: number of hops required for the backup path; Police[id]: stores the warning value of the next hop node.
[0045] The specific process of route generation includes:
[0046] In the first step, the node deletes all entry records in the routing table. In the source code, the NetworkEmptyForwardingTable(node, BM-OLSR) function is used to clear all routing entries.
[0047] The second step is to generate the following routing table entries from the node's one-hop neighbor table (h=1) and add them to the routing table;
[0048] R_dest_addr = 1 – hop neighbor node address;
[0049] R_next_addr = 1 – hop neighbor node address;
[0050] R_dist = 1;
[0051] Step 3: Generate the following routing table from the node's two-hop neighbor table (h=2) and add it to the routing table;
[0052] R_dest_addr=the address of the 2-hop neighbor;
[0053] R_next_addr=the R_next_addr of the entry in the routing table with;
[0054] R_dest_addr=N_neighbor_addr of the 2–hop tuple;
[0055] R_dist = 2;
[0056] Step 4: Add the routing table entry of the destination node with a distance of (h+1) to the routing table, starting from h=2, for each entry in the local node topology table,
[0057] if((no entry that R_dest==T_dest in routing table)&&(have exist theentry thar R_dest==T last))
[0058] / *If there is no one-hop entry in the routing table with R_dest_addr equal to T_dest_addr, and there is an entry in the routing table with R_dest_addr equal to T_last_addr, then add a new entry to the routing table: * /
[0059] R_dest_addr = T_dest_addr;
[0060] R_next_addr=R_next_addr of the recorded route entry where;
[0061] R_dest_addr = T_last_addr;
[0062] R_dist=h+1;
[0063] Step 5: Based on step 4:
[0064]
[0065] At this point, the backup path is established. Each drone node can establish a complete routing table. The so-called ideal situation here is that each node has understood the topology of the entire network through HELLO messages and TC control messages.
[0066] The OLSR routing protocol is an a priori routing protocol. The routing protocol proposed in the present invention should also be a priori when switching the backup route, rather than waiting until the route is detected to be invalid or the link is disconnected before switching the backup route, which would take too long to reconstruct the route. Therefore, a timing function is set up to periodically check the link status and preventively switch the backup route based on the result of mobility prediction.
[0067] like Figure 4As shown in the figure, in the ns-3 model, the arrow represents the trajectory of the drone node movement. Node 3 is set to leave along the trajectory of the arrow. At this time, the mobile prediction model predicts that it is beyond 80% of the transmission range of node 1 at a certain moment. Therefore, node 3 is predicted to be an unstable node relative to node 1. Once the periodic inspection timing function is triggered, the original main path 1->3->5->6 from node 1 to node 6 will switch to the backup path 1->2->4->6 as the main path (other nodes are not considered for the time being). After the backup path is switched to the main path, the protocol no longer searches for a new backup path, but only when the backup path fails, the protocol will re-initiate the routing search process.
[0068] Due to the limited payload of drones, the batteries they carry cannot be too large, so they rely heavily on energy. The MPR nodes of the OLSR protocol frequently forward TC (topology control) packets even when there is no communication. If a node has little remaining energy, it can only maintain its own flight and cannot continue to send data. If it is selected as an MPR node again, its energy will be exhausted quickly, and the entire route will be broken, which is not conducive to drone self-organizing networks. Therefore, in order to ensure the fairness of energy consumption, the remaining energy of the node should also be taken into account when selecting the MPR node.
[0069] Inputting the node information into the node mobility prediction model, the mobility prediction result of the node is obtained, including: the node information includes the remaining energy of the node, the link failure time and the signal strength of the node;
[0070] The residual energy of the node is input into the node mobility prediction model based on the residual energy for mobility prediction; if the predicted stability of the neighboring node is less than zero, a warning value is set for the neighboring node, and if the predicted stability of the neighboring node is greater than zero, the neighboring node is used as a switching route; the Gauss-Markov mobility model is used to obtain the position and movement information of two adjacent nodes, and the link failure time between the two nodes is calculated based on the position and movement information of the adjacent nodes; the link failure time between the two nodes is input into the node mobility prediction model based on the link failure time for mobility prediction; when the link failure time of the two adjacent nodes is predicted to be less than or equal to zero, a warning threshold is set for the neighboring node, otherwise the neighboring node is used as a switching route;
[0071] Obtain the received packets received by each node, and detect the power of the sending node based on the received packets; predict the mobility of the node based on the power of the sending node; if the predicted stability of the neighboring node is less than zero, set a warning value for the neighboring node; if the predicted stability of the neighboring node is greater than zero, use the neighboring node as a switching route.
[0072] When a node is selected as an MPR set, if the node has insufficient residual energy, it should not be selected as an MPR node even if the connection is relatively large. Therefore, it is necessary to measure the residual energy of the node. Let the initial energy of node i be E0, and the energy consumed in one routing be E i , then the current remaining energy of node i is:
[0073] E r (i) = E0-E i
[0074] Where: E0 can be set initially, E i It can be calculated.
[0075] Assume that the power of receiving data of node i is P1, the power of sending data is P2; the duration of receiving data packet is T1, the duration of sending data packet is T2; the length of receiving data packet is L1, the length of sending data packet is L2; the data receiving rate is V1, the data sending rate is V2, then the energy consumed by node i in this route is:
[0076] E i =P1T1+P2T2=P1L1 / V1+P2L2 / V2
[0077] Among them, P1 represents the power of node i receiving data, T1 represents the duration of receiving data packets, L1 represents the length of receiving data packets, V1 represents the data receiving rate, P2 represents the power of node i sending data, T2 represents the duration of sending data packets, L2 represents the length of sending data packets, and V2 represents the data sending rate.
[0078] The Gauss-Markov mobility model can predict the location of neighbor nodes at any time, but it is not reliable to select the next-hop node based solely on geographic location information. The high-speed mobility of drone nodes makes the link unstable and prone to breakage. Therefore, the link expiration time LET (Link Expiration Time, LET) is introduced to predict the stability of drone links. It is assumed that all nodes in the drone network are transmitted in a free space propagation model, the signal strength is only related to the propagation distance, each node has the same signal transmission capability and propagation range, and each node is synchronized using the network time protocol or GPS system. Therefore, it is possible to consider that two nodes within the effective transmission range are always connected.
[0079] The use of GPS system provides the location and movement information of nodes to calculate the link failure time (LET) between any two nodes, such as Figure 2 As shown, two nodes m and n are within each other’s transmission range. (x m ,y m ) and (x n ,yn ) are the coordinates of nodes m and n respectively. m and v n is the average velocity of nodes m and n. θ m and θ n (0<θ m ,θ n <2π) is the moving direction of the node. When two nodes move along the straight line in their direction, the transmission radius r between the two nodes after time t. If the nodes continue to move, their distance will be greater than the transmission radius r, and the transmission link between them will be interrupted. The transmission radius can be expressed as:
[0080] r 2 =(aLET+b) 2 +(cLET+d) 2
[0081] a=v m cosθ m -v n cosθ n
[0082] b = x m -x n
[0083] c=v m sinθ m -v n sinθ n
[0084] d=y m -y n
[0085] Where LET represents the link failure time, a represents the average speed difference between node m and node n in the horizontal direction, b represents the position distance between node m and node n in the horizontal direction, c represents the average speed difference between node m and node n in the vertical direction, d represents the position distance between node m and node n in the vertical direction, and v m represents the average speed of node m, v n represents the average speed of node n’s movement, θ m represents the moving direction of node m, θ n represents the moving direction of node n, x m Represents the horizontal coordinate of node m, x n Represents the horizontal coordinate of node n, y m Represents the ordinate of node m, y n Represents the vertical coordinate of node n.
[0086] The time t is called the link failure time (LET) between two nodes and can be expressed as:
[0087]
[0088] If v m =v n And θ m =θ n , then the moving speed and direction of the two nodes are consistent. When LET→∞, it can be judged that the two adjacent nodes are always connected and tend to be stable; when LET<0, the two nodes are not connected and the two nodes are unstable. In summary, only when LET>0, the two neighboring nodes tend to be stable, and the larger the LET, the stronger the stability of the node.
[0089] In wireless ad hoc networks, due to the mobility of nodes, nodes can detect the power of the node that sends the packet by receiving packets. Because each node has a unique identifier, the node that receives the packet can quickly know who sent the packet. In this way, through the change in power, we can determine whether the neighboring node is far away from the source node. If nodes A and B are neighbors, node A can derive the relative movement speed of the two nodes based on the power intensity of the packet sent by node B. The formula is as follows:
[0090]
[0091] in, Indicates the receiving power of node A to node B at the current moment; It indicates the received power of node A from node B at the last moment. From the above formula, it can be concluded that if V AB ≥0, indicating that node A and node B are close to each other or keep the distance unchanged, and the stability of the two nodes is strong; otherwise, it means that node A and node B are far away from each other, and the stability between the two nodes is weak.
[0092] Each node records the signal strength, remaining energy, and link failure time of the received broadcast packet at the physical layer. If the signal strength exceeds the threshold or the remaining energy can only sustain the flight and the link failure time is less than 0, it is recorded as a warning police[NodeId]. The same is true for other nodes.
[0093] The node periodically detects the police value of the next-hop route in the routing table, and decides whether to switch based on the police value and the backup route. Figure 5 shown.
[0094] like Figure 6As shown in the figure, the comparison of the packet delivery rate of nodes under different moving speed conditions. It can be seen from the figure that as the node moving speed increases, the time it takes for the node to move to the pre-set destination will become shorter. Therefore, if the entire simulation time remains unchanged, the node will run to the destination faster, thereby shortening the time it takes for the network link to break, and the system will become relatively stable. Therefore, from Figure 6 It can also be seen that as the node moving speed increases, the packet delivery rate also increases. When the speed is high, the present invention is better than the original protocol, which shows that when the node moves at high speed, the backup route established by the present invention plays a role.
[0095] like Figure 7 As shown, under lighter loads, the protocol proposed in the present invention needs to predict node information, and the network congestion level is higher than the original protocol. However, when the speed increases, the network topology changes more sharply, and the original protocol will generate more routing update information. However, due to the existence of backup routes, the improved protocol has lighter network congestion than the original protocol, and the end-to-end delay is also reduced.
[0096] like Figure 8 As shown, when the network load is light and the speed is low, the jitter of the original protocol and the improved protocol are relatively small. At low speed, since the present invention needs to record the signal strength of the node, the remaining energy of the node, the location information of the node, etc., the effect is not as good as the original. However, when the moving speed becomes faster, the topology changes faster. The original protocol requires a lot of route reconstruction delay due to frequent route breaks, resulting in a higher jitter of the original protocol than the improved protocol.
[0097] like Fig. 9 As shown in the figure, as the network load increases, the routing overhead of the improved protocol increases relative to the original protocol. When the node speed is low, the network topology changes slowly, and the routing update packets are relatively small, the routing overhead of the improved protocol is higher than that of the original protocol. However, when the network topology changes quickly, the original protocol maintains the route by constantly updating the routing table, while the improved protocol has a backup route, so the number of times the route is re-initiated is reduced, and the routing overhead is reduced.
[0098] The present invention conducts relevant research on the mobility prediction routing protocol of UAV networking. In this invention, the problem of frequent disconnection of communication links caused by factors such as highly dynamic changes of UAV nodes, limited node energy, and limited node communication distance is solved. A design method of mobility prediction multipath OLSR routing protocol based on UAV networking is proposed, which takes into account the characteristics of the above UAV nodes and the problem of frequent routing disconnection. A primary-backup routing mechanism is adopted, and a backup route is established on the original route, and three mobility prediction mechanisms are added to predict the remaining energy, communication distance, and link survival time of UAV nodes, and timely switch the backup route according to the prediction results. Simulation results show that the scheme in this paper has lower routing overhead and better robustness.
[0099] The above embodiments further illustrate the purpose, technical solutions and advantages of the present invention in detail. It should be understood that the above embodiments are only preferred implementation modes of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made to the present invention within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks, characterized in that: include: Establish a drone mobile node system model based on the drone's moving path; According to the constructed UAV mobile node system model, a multipath routing algorithm is used to establish a primary backup route, where the primary backup route includes a primary path and a backup path; the information of the node to be predicted is obtained, and the node information is input into the node mobility prediction model to obtain the mobility prediction result of the node, and the route switching is performed according to the prediction result; The process of inputting node information into the mobility prediction model includes: the node information includes the node's remaining energy, link failure time, and the node's signal strength; The residual energy of the node is input into the node mobility prediction model based on the residual energy to perform mobility prediction; if the predicted stability of the neighboring node is less than zero, a warning value is set for the neighboring node; if the predicted stability of the neighboring node is greater than zero, the neighboring node is used as a switching route; The Gauss-Markov mobility model is used to obtain the location and mobility information of two adjacent nodes, and the link failure time between the two nodes is calculated based on the location and mobility information of the adjacent nodes; the link failure time between the two nodes is input into the node mobility prediction model based on the link failure time to perform mobility prediction; when the link failure time between the two adjacent nodes is predicted to be less than or equal to zero, a warning threshold is set for the neighbor node, otherwise the neighbor node is used as a switching route; Obtain the received packets received by each node, and detect the power of the sending node based on the received packets; predict the mobility of the node based on the power of the sending node; if the predicted stability of the neighboring node is less than zero, set a warning value for the neighboring node; if the predicted stability of the neighboring node is greater than zero, use the neighboring node as a switching route.
2. According to claim 1, a multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks is characterized in that: The process of establishing the primary and backup routes using the multipath routing algorithm includes: establishing multiple available routes between the source node and the destination node, and taking the route with the best link quality as the primary route; based on the remaining energy of the node, using the Gauss-Markov mobility model to predict the node's position at the next moment, when it is predicted that the neighboring node is out of communication range or the link is about to fail, switching to the backup route in advance.
3. According to claim 1, a multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks is characterized in that: The process of obtaining the remaining energy of a node includes: setting the initial energy E0 of the node, calculating the energy E0 consumed by the node to perform a routing operation i , the remaining energy of the node is calculated based on the initial energy and the consumed energy; the formula for the energy consumed by the node to perform a routing operation is: <h2 style=";text-align:left;direction:ltr">E<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> (P1T1+P2T2) (P1L1 / V1+P2L2 / V2) Among them, P1 represents the power of node i receiving data, T1 represents the duration of receiving data packets, L1 represents the length of receiving data packets, V1 represents the data receiving rate, P2 represents the power of node i sending data, T2 represents the duration of sending data packets, L2 represents the length of sending data packets, and V2 represents the data sending rate.
4. The multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks according to claim 1 is characterized in that: The process of calculating the link failure time between two nodes includes: using the GPS system to obtain information of the two nodes m and n, the information including the coordinates of the nodes, the average speed of the node movement, the moving direction of the node and the transmission radius r of the two nodes; calculating the link failure time based on the information of the two nodes.
5. A multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks according to claim 4, characterized in that: The formula for calculating link failure time is: a=v m cosθ m -cosθ n b=x m -x n c=v m sinθ m -v n sinθ n d=y m -y n Where LET represents the link failure time, a represents the average speed difference between node m and node n in the horizontal direction, b represents the position distance between node m and node n in the horizontal direction, c represents the average speed difference between node m and node n in the vertical direction, d represents the position distance between node m and node n in the vertical direction, and v m represents the average speed of node m, v n represents the average speed of node n’s movement, θ m represents the moving direction of node m, θ n represents the moving direction of node n, x m Represents the horizontal coordinate of node m, x n Represents the horizontal coordinate of node n, y m Represents the ordinate of node m, y n Represents the vertical coordinate of node n.
6. The multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks according to claim 1 is characterized in that: The mobility prediction of the node based on the power of the packet-sending node includes: obtaining the receiving power of node A to node B at the current moment and the received power received by node A from node B at the last moment According to the obtained and Calculate the relative motion speed V of the two nodes AB , when V AB When ≥0, node A and node B are close to each other or keep the distance unchanged, and the stability of the two nodes is strong; otherwise, node A and node B are far away from each other, and the stability between the two nodes is weak.
7. A multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks according to claim 6, characterized in that: The formula for relative motion speed is: in, Indicates the receiving power of node A to node B at the current moment; Indicates the received power that node A received from node B at the last moment.
8. The multipath routing protocol method for mobility prediction of unmanned aerial vehicle ad hoc networks according to claim 1 is characterized in that: The node performs route switching including periodically sending HELLO messages and TC packets containing predicted location information to detect neighbor nodes in the route list. When the detected route warning value of the neighbor node is 1, the backup route is enabled, otherwise the backup route is not enabled.
