Dynamic routing method for high-maneuverability flight ad hoc network

Abstract

The invention discloses a three-dimensional space high-mobility ad hoc network routing method based on a multi-beam digital array antenna, and belongs to the field of wireless ad hoc network communication. The aerial high-mobility node adopts a high-gain digital array multi-beam antenna, and can establish a high-speed communication link with a plurality of surrounding nodes at the same time. A node obtains a neighbor table of the node through neighbor interception, selects a most stable MPR node set with the minimum overhead through an optimized MPR selection mechanism, forwards other nodes through TC grouping, recovers a network topology structure after obtaining all TC grouping information in a network, calculates a main path and a plurality of standby paths through routing selection measurement factors, and forwards the main path and the standby paths to a network. And when the main link state goes wrong, the alternative path can be quickly switched, so that the reliability and low time delay of the network are effectively ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless ad hoc network communication, and particularly relates to a dynamic routing method for a highly mobile flight ad hoc network. Background Art

[0002] A Mobile Ad hoc Network (MANET) is a self-organizing network without a fixed infrastructure. Nodes can move freely in a wireless environment and dynamically form a network. When the nodes are unmanned aerial vehicles or other aircraft, this self-organizing network is also called a Flying Ad hoc Network (FANET). This type of self-organizing network requires an efficient routing protocol to handle dynamic topology changes and multi-hop transmissions. OLSR (Optimized Link State Routing) is a routing protocol designed for this type of self-organizing network, mainly used in networks with dynamic topologies. By optimizing the propagation of link state information, it improves the efficiency of route selection and the overall performance of the network.

[0003] The core mechanism of OLSR is the Multipoint Relay (MPR) mechanism to reduce the number of broadcast messages. Each node selects a part of its neighbor nodes as relay nodes, which are responsible for forwarding link state messages, thereby reducing the network load. The algorithm for selecting multipoint relay nodes in the OLSR protocol mainly includes two steps: 1. Select the first-hop neighbors that cover and isolate the second-hop neighbors, mainly to improve data transmission efficiency and network reliability, avoid packets passing through too many hops, and reduce the possibility of delay and packet loss; 2. Select other first-hop neighbors according to the maximum coverage standard to ensure that each node in the network can be fully covered and connected, expanding the scope and stability of the network.

[0004] Due to the frequent link state updates of the Link State Routing (OLSR) protocol, to a certain extent, it leads to an increase in the overhead of control information, and it is also not very suitable for the highly mobile FANET application scenario. In the case of rapid node movement, OLSR may be difficult to maintain an effective route.

[0005] Unmanned aerial vehicle nodes move at high speed in a wide-area three-dimensional space, and the set of adjacent nodes is constantly changing. The network routing table also needs to be updated in real time accordingly. Facing such a rapidly changing topology, traditional routing algorithms cannot be directly applied. Compared with other routing protocols, proactive routing protocols are more suitable for networks with high requirements for real-time performance and reliability. However, proactive routing algorithms usually introduce a large amount of network overhead. How to reduce network overhead is a very critical issue for the FANET communication system.

[0006] Therefore, it is necessary to design a highly reliable and low-latency routing algorithm, focusing on real-time performance and reliability indicators. Summary of the Invention

[0007] Aiming at the technical problems that the OLSR protocol has a slightly large overhead and cannot adapt to the high - mobility FANET application scenario, the present invention provides a dynamic routing method for a highly mobile flying ad - hoc network. Nodes communicate with each other through a digital array multi - beam antenna for self - organizing networking. Each node can simultaneously establish directional communication links with multiple neighbor nodes, and at the same time, the routing table is updated in real time through a high - dynamic routing algorithm. Through routing relay, fast and reliable communication between all nodes in the network is realized, which is applicable to the FANET scenario where UAV nodes move at high speed and have limited energy and bandwidth.

[0008] The dynamic routing method for a highly mobile flying ad - hoc network protected by the present invention optimizes the OLSR protocol. The main process of the entire routing protocol is as Figure 1 shown.

[0009] I. Link Awareness

[0010] At the beginning of the self - organizing network establishment, each node performs link awareness, that is, each node obtains its one - hop and two - hop neighbor information through neighbor listening and records it in the one - hop neighbor table and the two - hop neighbor table.

[0011] The way to obtain the two - hop neighbors of a node is that all nodes send their own one - hop neighbor tables to their neighbor nodes, so that all nodes can obtain all their two - hop neighbor information.

[0012] The information of the one - hop neighbor nodes included in the one - hop neighbor table are: local node, link status, forwarding willingness, one - hop neighbor, as shown in Table 1.

[0013] Table 1

[0014] Local node Link state Forwarding willingness One-hop neighbors of this node

[0015] The information included in the two - hop neighbor table are: the number of two - hop neighbors, each two - hop neighbor node and its relay node, as shown in Table 2.

[0016] Table 2

[0017]

[0018] II. MPR Selection

[0019] Calculate the MPR node set of each node according to the one - hop and two - hop neighbor tables of each node and record it in the MPR table of this node (see Table 3). The MPR node set of each node consists of some of its one - hop neighbors.

[0020] Table 3

[0021]

[0022] The MPR selection criterion of the present invention is that the local node can reach all two-hop neighbor nodes through MPR nodes, and the fewer the number of MPR nodes, the better; the MPR selection principle is to preferentially select nodes with a high MPR metric factor, and the MPR metric factor is a weighted expression of four indicators: the coverage node degree, the node forwarding willingness, the link stability degree, and the node load condition.

[0023] The MPR selection algorithm provided by the present invention includes the following steps:

[0024] 1. Select one-hop neighbor nodes that can cover isolated two-hop neighbors and add them to the MPR node set, where an isolated two-hop neighbor refers to a node that can only be connected to the local node through a unique neighbor node;

[0025] 2. Traverse the remaining two-hop neighbor nodes (two-hop neighbors other than isolated two-hop neighbors), find the corresponding one-hop neighbor nodes. If there are multiple one-hop neighbor nodes for a two-hop neighbor, first select as few MPR nodes as possible based on the MPR selection criterion. Otherwise, select a node with a higher MPR metric factor and add it to the MPR node set until the local node can cover all two-hop neighbors.

[0026] III. TC packet message processing

[0027] The TC packet message processing flow (taking node P as an example) is as Figure 2 shown and includes the following steps:

[0028] 1. When the MPR nodes of node P are updated, recreate the TC packet, update the TC sequence number (increment by one), and forward it to all its neighbors. The TC packet message is shown in Table 4.

[0029] Table 4

[0030] TC sequence number Hop count Source node Relay node Number of MPRs Neighbor 1 Neighbor 2 ... Neighbor k

[0031] 2. When the neighbor of node P receives the TC packet, first compare it with the original TC packet message to determine whether it is the latest TC packet message for this node (the one with a larger TC sequence number is the new TC packet message). If so, update the topology table (Table 5) and the MPR neighbor table (see Table 6), and then determine whether to forward it. If forwarding is required, send it to all its neighbors; otherwise, discard the data packet. Each node records the TC packet information that needs to be forwarded in the TC message table (refer to Table 7).

[0032] Table 5

[0033]

[0034] Table 6

[0035] MPR Neighbor 1 MPR Neighbor 2 MPR Neighbor 3 MPR Neighbor 4 …

[0036] Table 7

[0037] Node TC sequence number Neighbor 1 Neighbor 2 ... Neighbor k

[0038] The judgment principle of whether to forward is as follows: 1. Whether itself is the MPR node of the neighbor node. If so, forward. Whether the current forwarding times of the TC packet reach the upper limit. If they reach the upper limit, do not forward.

[0039] 3. Through layer-by-layer forwarding, each node can obtain the TC packet messages of other nodes in the network, and then restore the whole network topology structure. Then, according to the topology structure, calculate the routes from the local node to other nodes. Each node maintains a routing table for saving the routes to each destination node, as shown in Table 8.

[0040] Table 8

[0041]

[0042] IV. Routing calculation

[0043] Since there may be more than one, or even several, routing paths from the source node to the destination node. The source node selects the routing path through the routing metric factor, where the routing metric factor depends on the hop count, link inventory time, load condition of the relay node, and link rate. The present invention recommends calculating 1 main path and n alternative paths (n is preferably 2). Before route selection, first judge whether the main path is available (whether the node distance exceeds the maximum communication radius). If the main path is available, select the main path. Otherwise, perform the alternative path judgment. If none of them are available, the current route fails.

[0044] The present invention provides the following routing table calculation method (taking node Q as an example):

[0045] ⑴ Clear the routing table of node Q.

[0046] ⑵ Add all one-hop neighbors M1 of node Q to the routing table.

[0047] ⑶ Traverse all two-hop nodes M2 of node Q, find the corresponding one-hop relay node M1, add the non-repeated two-hop neighbors to the routing table, and record its previous hop node as M1; if there are multiple paths, record them in the routing table from high to low according to the routing metric factor.

[0048] ⑷ Traverse all three-hop nodes M3 of node Q. If there is a node that has not been added to its routing table, add it to the routing table and look up its one-hop node in the table; if it is a node that has already been added to the routing table, determine whether the one-hop node of the current path is the same as that of the existing path. If they are the same, do not add it. Otherwise, calculate the routing metric factor and record it in the routing table in descending order of the metric factor. To avoid recording too many routing paths and causing resource waste, it can be set to only record the top three groups of paths with the highest routing metric factors.

[0049] ⑸ Traverse all four-hop, five-hop,..., N-hop nodes of node Q in sequence, and the operations are the same as in step ⑷.

[0050] V. Routing Table Maintenance

[0051] The messages of OLSR are updated periodically, and each message has a certain validity period. Assume that a link in the network is interrupted, resulting in a node loss. After updating the neighbor messages subsequently, through the hello message and TC message updates, the network needs to recalculate the routing table.

[0052] Advantages of the present invention:

[0053] 1. Through the proactive routing method, the routing table is updated in a timely manner when the topology changes, ensuring the communication reliability and delay performance under high-dynamic topology changes.

[0054] 2. In the traditional MPR selection strategy, a link stability index is introduced, which improves the validity period of MPR nodes and effectively reduces the network overhead.

[0055] 3. An MPR node judgment is added to the TC packet. Only MPR nodes can forward. Compared with the processing method in the traditional TC packet where all neighbors forward, it helps to restore the network topology information and also reduces the network overhead.

[0056] 4. Based on the traditional routing table calculation strategy, first, the routing metric factor is optimized. In addition to the number of hops, factors such as link survival time, link rate, and relay node load are introduced, which extends the validity period of the routing table. In addition to the main path, two alternative paths are additionally recorded to avoid link failures introduced by high-mobile topologies; at the same time, the path validity is judged at each moment, effectively improving the reliability of the network routing performance. Description of the Drawings

[0057] Figure 1 It is a flowchart of the dynamic routing method;

[0058] Figure 2 It is the processing flow of the TC packet message;

[0059] Figure 3 It is the network topology relationship in Embodiment 1. Detailed Embodiments

[0060] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are given for purposes of illustration and description, and are not exhaustive or limit the present invention to the disclosed form. Many modifications and variations will be obvious to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and enable those of ordinary skill in the art to understand the present invention and thus design various embodiments with various modifications suitable for specific purposes.

[0061] Embodiment 1

[0062] In this embodiment Figure 3 Taking the network topology relationship shown as an example, the self-organizing network dynamic routing method proposed by the present invention will be specifically described. Figure 3 There are six nodes A, B, C, D, E, and F in the shown network. Hereinafter, node A will be used as the local node for explanation.

[0063] I. Link Awareness

[0064] Node A obtains one-hop neighbor information through neighbor listening and records it in the one-hop neighbor table (Table 9). As Figure 3 can be seen, node A has four (one-hop) neighbor nodes, namely node B, node G, node H, and node F; while node B has three (one-hop) neighbor nodes, namely node A, node H, and node E. Similarly, the (one-hop) neighbor nodes of node G, node H, and node F can also be known.

[0065] Table 9 One-hop Neighbor Table of Node A

[0066] Neighbor node Link state Forwarding willingness Neighbor 1 Neighbor 2 Neighbor 3 Neighbor 4 B 1 1 A H E G 2 2 A H C F H 2 1 A B G F 3 1 A G C E

[0067] All nodes send their own one-hop neighbor tables to their (one-hop) neighbor nodes, so that all nodes can obtain all their two-hop neighbor information and record it in the two-hop neighbor table (Table 10).

[0068] Node A has two two-hop neighbors, namely node E and node C. And there are two paths from node A to node E, namely A→B→E and A→F→E. There are also two paths from node A to node C, namely A→G→C and A→F→C.

[0069] Table 10 Two-hop Neighbor Table of Node A

[0070]

[0071] II. MPR Selection

[0072] Due to the MPR selection criteria: 1. The local node can reach all two-hop neighbor nodes through the MPR nodes; 2. The fewer MPR nodes, the better.

[0073] To meet criterion 1, first, the one-hop neighbor nodes that can cover the isolated two-hop neighbor nodes need to be added to the MPR node set to ensure that the local node can reach these isolated two-hop neighbor nodes. Since node A has no isolated two-hop neighbor nodes, there is no need to consider it.

[0074] Node A has two two-hop neighbor nodes, and there are three relay nodes, namely node B, node G, and node F, to reach these two two-hop neighbor nodes. However, based on criterion 2, only node F needs to be selected among the three, because through the relay node F, the two two-hop neighbor nodes of node A can be reached.

[0075] Therefore, the MPR table of node A is shown in Table 11, which includes the numbers of each MPR node and the valid time of this table.

[0076] Table 11 MPR Table of Node A

[0077] MPR table valid time MPR1 100 F

[0078] III. TC Packet Message Processing

[0079] Node A creates a TC packet message (Table 12) and sends it to all its neighbors (node B, node G, node H, node F).

[0080] Table 12

[0081] TC sequence number Hop count Source node Relay node Number of MPRs Neighbor 1 Neighbor 2 Neighbor 3 Neighbor 4 1 1 A * 1 F G H B

[0082] Since the only MPR node of node A is node F, only node F forwards this TC packet message to all its neighbors (node G, node C, node E).

[0083] At the same time, node A will also receive TC message packets created by other nodes. The forwarding process is omitted here. There are two possibilities for the TC packet message created by node E received by node A, as shown in Table 13 or Table 14 respectively. Based on this, two routing paths from node A to node E are obtained.

[0084] Table 13

[0085] TC sequence number Hop count Source node Relay node Number of MPRs Neighbor 1 Neighbor 2 Neighbor 3 1 2 E B 2 B F D

[0086] Table 14

[0087] TC sequence number Hop count Source node Relay node Number of MPRs Neighbor 1 Neighbor 2 Neighbor 3 1 2 E F 2 F C E

[0088] After receiving all the TC packet messages, node A restores the network topology table (Table 15).

[0089] Table 15

[0090]

[0091]

[0092] IV. Routing Calculation

[0093] 1. Add the one-hop neighbor M1 of node A to its routing table, see Table 16.

[0094] Table 16

[0095]

[0096] 2. Traverse the two-hop nodes M2 of node A, find the corresponding one-hop relay node M1, add the non-duplicate two-hop neighbors to the routing table, and record their previous hop node as M1. If there are multiple paths, record them in the routing table from high to low according to the routing metric factor, see Table 17.

[0097] Table 17

[0098]

[0099] 3. Traverse the three-hop nodes M3 of node A. There is a node D that has not been added to the routing table, and there are two paths from node A to node D: A→G→C→D and A→B→E→D. Therefore, add it to the routing table and record its two relay nodes (node C and node E), see Table 18.

[0100] In the path A→H→G→C, node C is a three-hop node of node A; in the path A→G→F→E, node E is also a three-hop node of node A. Since both node C and node E have been added to the routing table of node A, further judgment is needed, that is, to judge whether the one-hop nodes of the current path are the same as those of the existing paths.

[0101] Combined with Figure 3 , for the path A→H→G→C, C is a three-hop node and has no common one-hop nodes with the two paths A→F→C and A→G→C. Therefore, it is also recorded in Table 18.

[0102] Combined with Figure 3 , for the path A→G→F→E, E is a three-hop node and has no common one-hop nodes with the two paths A→F→E and A→B→E. Therefore, it is also recorded in Table 18.

[0103] Table 18

[0104]

[0105] 4. Traverse the four-hop nodes M4 of node A. At this timeFigure 3 There are no nodes in the shown network topology that have not been added to the routing table of node A.

[0106] In the path A→H→G→F→E, node E is a four-hop node of node A, and it has no common one-hop node with the paths A→F→E, A→B→E, and A→G→F→E. Therefore, it is also recorded in the routing table of node A. See Table 19.

[0107] In the path A→G→C→D→E, node E is also a four-hop node of node A, but it has a common one-hop node with the path A→G→F→E. Therefore, it is not recorded.

[0108] Table 19

[0109]

[0110]

[0111] 5. Traverse the five-hop node M5 of node A. In the path A→H→G→F→E→D, node D is a five-hop node of node A, and it has no common one-hop node with A→G→C→D and A→B→E→D. Therefore, it is also recorded in the routing table of node A. See Table 20.

[0112] Table 20

[0113]

[0114] V. Routing Table Maintenance

[0115] When the neighbor changes or the TC message is updated, recalculate the routing table.

[0116] Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art and related fields based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

Claims

1. A dynamic routing method for a high-maneuverability flight ad hoc network, wherein nodes communicate with each other through digital array multi-beam antennas, and each node can establish directional communication links with multiple neighboring nodes at the same time, characterized in that: The following steps are involved: Step 1: At the beginning of the self-organizing network establishment, each node is allowed to perform link perception, that is, each node obtains its one-hop and two-hop neighbor information through neighbor listening, and records it in the one-hop neighbor table and the two-hop neighbor table; The one-hop neighbor table contains the following information about the one-hop neighbor node: local node, link status, forwarding intention, and one-hop neighbor. The two-hop neighbor table contains the following information: the number of two-hop neighbors, each two-hop neighbor node and its relay node. Step 2: Each node calculates the MPR node set based on its own one-hop and two-hop neighbor tables and records it in the MPR table. The MPR node set of each node consists of some of its one-hop neighbors. The MPR selection criterion is that the local node can reach all two-hop neighbor nodes through the MPR node, and the fewer the number of MPR nodes, the better; the MPR selection principle is to give priority to nodes with high MPR measurement factors, which are weighted expressions of four indicators: coverage node degree, node forwarding willingness, link stability, and node load conditions; Step 3: Each node creates TC group information and sends it to all neighbors. The TC group information includes the TC sequence number, hop count, source node, relay node, MPR number, and all neighbor nodes. The neighbors that receive the TC group message determine whether they are the MPR node of the node. Only the MPR node will continue to forward the TC group message to its neighbors, and so on. After forwarding layer by layer, each node can obtain the TC group message of other nodes in the network, and then restore the topology of the entire network, and then calculate the route from the local node to other nodes based on the topology. Step 4: The source node selects a routing path using a routing metric factor, where the routing metric factor depends on the number of hops, link inventory time, relay node load, and link rate.

2. An electronic device, characterized in that: It includes a memory, a processor and a bus; the bus is used to realize the connection and communication between the memory and the processor; the processor is used to execute the computer program stored in the memory; when the processor executes the computer program, the steps in the dynamic routing method of high-maneuverability flight self-organizing network described in claim 1 are realized.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the high-maneuverability flight ad hoc network dynamic routing method described in claim 1 are implemented.

Images