Node backup method for lunar man-machine joint exploration network

Abstract

The present invention discloses a node backup method for a lunar man-machine joint exploration network, and belongs to the field of satellite network communication technology. The method of the present invention first combines the particularity of lunar missions and their networking methods to model the lunar man-machine joint exploration network, classify the actors in the network, and constrain the node behavior. In order to reduce the information overhead of each node maintenance, the node determines the critical point through two-hop neighbor information. The critical point uses one-hop neighbor information and Hello message information to select backup nodes based on the principles of node type, recovery overhead, and the degree of impact on the network. The independence of node tasks is considered during fault recovery, and a fault transfer strategy is used to offset the mobility responsibility of astronaut nodes. Experiments show that the node backup mechanism can generate a backup recovery path for each key node when there are redundant nodes in the network, and can reconnect the network without moving the astronaut nodes.

Description

Technical Field

[0001] The present invention belongs to the field of satellite network communication technology, and in particular relates to a node backup method for a lunar man-machine joint detection network. Background Art

[0002] The joint exploration of man and machine on the lunar surface is regarded as one of the most effective ways to fully explore the moon. The joint exploration network can achieve rapid and efficient lunar movement through astronauts' direct assessment of the environment; use astronauts to make intelligent problem decisions and respond quickly; and complete more complex tasks through human-machine collaboration, and improve the level of system management and maintenance. Traditional network communications are usually based on fixed infrastructure, such as mobile networks supported by wired networks, but such networks have high deployment costs, difficult wiring, and are prone to disasters. It is not realistic to establish similar traditional communication networks in the early stages of lunar exploration. Wireless multi-hop networking is one of the best ways to network for future joint exploration of man and machine on the lunar surface. Using this networking method can allow nodes on the lunar surface to quickly establish connections, realize real-time data transmission and command control, and improve the efficiency of task execution. The downside is that the wireless multi-hop networking method increases the probability of disconnection of the entire network. When the energy of a node on the lunar surface is exhausted or a fault occurs, if the node is a global cut point, the network will be disconnected. For the joint human-machine exploration on the moon, network disconnection will not only affect the smooth execution of the exploration mission, but the loss of connectivity will also pose a threat to the lives of astronauts in the complex lunar environment. Therefore, how to ensure connectivity when a node fails is an urgent problem to be solved. Summary of the invention

[0003] In view of the above problems, the present invention proposes a node backup method for the lunar man-machine joint exploration network. First, a model is established for the lunar man-machine joint exploration network, and a distributed node backup mechanism is proposed. After the network is initialized, the mechanism will formulate a backup strategy for all possible global cut points in the network. When a cut point fails, the network is reconnected by relocating the nodes on the backup path. The backup mechanism of the present invention also takes into account the independence of each astronaut's mission in the backup plan, and all astronaut nodes will not be moved during backup and recovery.

[0004] The above purpose is achieved through the following technical solutions:

[0005] The node backup method for the lunar surface human-machine joint detection network of the present invention comprises the following steps:

[0006] S1. Construct a lunar surface human-machine joint exploration network model: define the lunar surface gateway as an anchor point, each astronaut node as a random mobile node, and the mobile device in the exploration mission as a controllable mobile node, form a network in a wireless multi-hop manner, and obtain the coordinates of the anchor point, random mobile node, and controllable mobile node in the network;

[0007] S2, in random mobile nodes and controllable mobile nodes, the k-hop cut point judgment algorithm is used to determine the two-hop critical point;

[0008] S3, after the node is determined to be a two-hop critical point, it selects a node from its one-hop neighbor as its backup node;

[0009] S4. When the two-hop critical node determines the backup, it will send a Commission message to its backup neighbor. The node that receives the Commission message will add the two-hop critical node that sent the Commission message to its own master node set and start monitoring the heartbeat. At the same time, the two-hop critical node that sends the Commission message also monitors whether its own backup is alive. If its own backup fails or performs other backup tasks, the two-hop critical node that sends the Commission message will return to step S3 to reselect the backup.

[0010] S5. The two-hop critical point with backup needs to maintain the backup node path cost , To repair the total distance that all nodes move for a failed node, when receiving a Hello message with a lower recovery cost, the node will send a Cancel message to its own backup node to cancel the backup plan and select a new backup node.

[0011] Furthermore, the anchor points, random mobile nodes, and controllable mobile nodes in the network described in step S1 are all equipped with omnidirectional antennas, and the maximum transmission radius of each antenna is , when the distance between two nodes is less than A two-way communication link is considered to be established between them.

[0012] Furthermore, the k-hop cut point judgment algorithm is used in step S2 to determine the two-hop critical point, specifically: first, if the node is itself If the node is a cut point of the neighbor subgraph, the node is defined as After the network is initialized, each node periodically sends a Hello broadcast message with a TTL (Time To Live) of 2 to its neighbors, which carries basic information including its own ID, location, and node type so that other nodes can generate a two-hop subgraph to determine whether it is a two-hop critical point.

[0013] Furthermore, the selection principle of selecting a node from its one-hop neighbor as its own backup node in step S3 is as follows:

[0014] Principle 1: The backup node cannot be a two-hop critical point and cannot be a randomly moved node;

[0015] Principle 2: If there are multiple nodes that meet Principle 1, select the node with the smallest node degree, because moving this node will have the least impact on other nodes;

[0016] Principle 3: If there are multiple nodes that meet Principle 2, select the node closest to itself;

[0017] For the two-hop critical points of the backup node selected that do not meet the principles one, two, and three, the type and critical state of the two-hop critical point will be ignored when selecting the backup. Every time a Hello message is received from a neighboring node, the neighboring node of the two-hop critical point is checked to see whether there is a backup node. If there is a backup node, the backup node will be regarded as the backup of the two-hop critical point.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] The present invention establishes a model for the lunar man-machine joint exploration network and proposes a distributed node backup mechanism. The node backup mechanism provides a feasible node failure recovery scheme for the lunar man-machine joint exploration network. The mechanism makes full use of the redundant nodes in the network and allows the network to recover from the partition caused by the failure as much as possible without affecting the exploration mission of each astronaut. When the network is initialized, a backup node is selected for each key node. When the key node fails, a shortest backup recovery chain is formed to reconnect the network. Experiments verify that the backup mechanism converges when there are redundant nodes in the network, and the success rate of the cut point backup can reach 100%. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The network model of lunar surface human-machine joint detection of the present invention;

[0021] Figure 2 For the present invention Example of critical point judgment;

[0022] Figure 3 This is an example of the network topology for the joint human-machine exploration of the lunar surface of the present invention;

[0023] Figure 4 This is the RN2 backup and repair process of the present invention. Figure 4 (a) shows the movement of node CN3 during the RN2 backup repair process. Figure 4 (b) shows the movement of node CN1 during the RN2 backup repair process;

[0024] Figure 5 This is the CN4 backup and repair process of the present invention. Figure 5 (a) shows the movement of node CN3 during the CN4 backup and repair process. Figure 5(b) shows the movement of node CN1 during the CN4 backup and repair process;

[0025] Figure 6 For the backup convergence process of the present invention, Figure 6 (a) shows the critical point determination during the backup convergence process; Figure 6 (b) shows the first round of backup during the backup convergence process; Figure 6 (c) shows the second round of backup during the backup convergence process; Figure 6 Middle (d) shows the backup completed during the backup convergence process;

[0026] Figure 7 It is the backup success rate of the global cut point under different network conditions of the present invention. Figure 7 (a) shows the backup success rate of global cut points under different network scales. Figure 7 (b) shows the global cutpoint backup success rate under different network flexibility;

[0027] Figure 8 It is the backup convergence time of all two-hop critical points in different network states of the present invention. Figure 8 (a) shows the backup convergence time under different network scales. Figure 8 (b) shows the backup convergence time under different network flexibility. DETAILED DESCRIPTION

[0028] 1. Implementation scenario description

[0029] The present invention first builds a lunar surface human-machine joint detection network model such as Figure 1As shown in the figure, the nodes involved in the network can be roughly divided into astronauts, lunar rovers, lunar gateways, etc. In the exploration mission, the lunar gateway is the key node for communicating with the lunar backbone network and is responsible for determining the operation location and operation radius. In a mission, the location of the gateway node is basically fixed, so this type of node is defined as an anchor node (AN) in the model; astronauts have the dominant power in the exploration mission and can make flexible decisions based on the exploration needs. Because the exploration mission of each astronaut node has certain independence and uncertainty, this type of node is defined as a random-walking node (RN) in the model; mobile devices such as lunar rovers play an important role as auxiliary tools for exploration missions and mobile communication relays. These devices are controlled by anchor nodes and astronaut nodes, so they are defined as controlled-mobile nodes (CN) in the model. At the same time, random-walking nodes (RN) and controlled-mobile nodes (CN) are collectively referred to as nodes, also collectively referred to as actors.

[0030] The lunar surface human-machine joint exploration network is networked in a wireless multi-hop manner. Each node is equipped with an omnidirectional antenna, and the maximum transmission radius of each antenna is , when the distance between two nodes is less than When the anchor point is , it is considered that a two-way communication link can be established between them. Except for the anchor point, other nodes can move, and the CN node can move at a constant speed according to the command. The research on node positioning in the ground network has become mature. Space satellite navigation and positioning systems such as GPS (Global Position System) and Beidou Satellite Navigation System can provide high-precision navigation services in the entire region. At present, the Earth-Moon navigation system is not perfect. In order to solve the problem of position loss caused by the lack of satellite signals, many studies use the relative position between nodes to estimate the coordinates of the nodes. These methods are mainly divided into ranging and non-ranging according to different measurement methods. The ranging algorithm uses the strength of the received signal between nodes, the time difference between signals, the angle of arrival and other information to calculate the position of the node. The non-ranging algorithm uses the connectivity of the network or the number of hops between nodes and other information to estimate the coordinates of the node. The above-mentioned node positioning algorithms can be implemented using existing technologies.

[0031] 2 Backup method description

[0032] The extreme temperature, tiny dust particles, and radiation on the lunar surface can damage equipment, and the electromagnetic activity in a vacuum environment is strong, so equipment on the lunar surface is more prone to single point failures. When certain key nodes such as cut points fail, it may cause the loss of communication links between multiple actors. Node failures in the network are unpredictable, and the ideas for solving this problem are mainly divided into active prevention and passive recovery. Active prevention is mainly to improve the fault tolerance of the network and keep each node When a critical node fails, The links can still remain connected, thus improving the network's ability to withstand destruction. Passive recovery refers to actively monitoring whether there are faulty nodes in the network. When a faulty node is detected and the faulty node affects the connectivity of the entire network, other nodes in the mobile network are used to reconnect the network.

[0033] Although keeping the network Redundancy can greatly improve the network's anti-destruction performance, but this method will greatly increase the number of nodes required to be deployed in the same area, and each node needs to maintain multiple connections, which increases resource consumption. For the lunar man-machine joint exploration network with limited equipment and equipment resources, it is necessary to keep the network The connection is not realistic. The lunar exploration mission requires the use of fewer devices to explore as much area as possible. Connecting the network requires not only a large number of nodes, but also If they are not connected, the range of activities of each device will be greatly restricted.

[0034] The passive repair method is to move redundant nodes to restore network connectivity when some nodes in the network fail and cause the network to lose connectivity. Because node failures are unpredictable, a spontaneous recovery method must be used. Passive topology repair is the most feasible solution from a resource perspective. When restoring network connectivity, the task independence of random-walking nodes (RN) should be considered, so RN nodes should not be used as controllable repair nodes. The time and location of RN node failures cannot be predicted. After the failure causes network partitions, the partitions cannot communicate with each other. Even if there is a central node, all nodes in the network cannot be used for fault recovery. Therefore, node fault recovery and backup should be deployed in each node in a distributed manner. When a leaf node fails, the connectivity of the network will not be affected. In order to avoid redundant recovery, the optimal strategy is to restore only when the cut point fails. However, accurate cut point identification requires topological information of the entire network, and these methods involve huge communication overhead. At present, most studies use local topological information and sacrifice a certain accurate cut point identification rate to reduce communication and maintenance overhead.

[0035] The k-hop cut point judgment algorithm is a local cut point detection algorithm. Each actor determines whether it is a local cut point based only on limited local information. First, define The concept of jumping critical point, that is, if the node is itself If it is a cut point of the neighbor subgraph, the node is called Jump critical point. Figure 2 The one-hop neighbor node of node C is , from the one-hop local topology of point C, the failure of point C will affect The local topology connectivity is determined by the node C, so node C is a one-hop critical node. Because C connects two leaf nodes A and B, C is also a global cut point. Similarly, nodes F and D are also one-hop critical points but not two-hop critical points. When there is a loop in the topology, the accuracy of the k-hop critical point will be greatly affected. For example, The loop is composed only when When it increases to 6, the global cut point on the loop can be accurately identified. When such a loop exists in the network, the route will be too long, so even if a misjudgment occurs, the topology can be reconstructed to shorten the information routing time. In order to balance resource overhead and recognition accuracy, the cut point identification strategy of the present invention uses two-hop topology information when judging critical nodes. After the network is initialized, each node periodically sends a Hello broadcast message with a TTL (Time To Live) of 2 to its surrounding neighbors, which carries its own ID, location, node type and other basic information so that other nodes can generate a two-hop subgraph to determine whether it is a two-hop critical point.

[0036] After a node determines that it is a two-hop critical point, the next step is to select a suitable backup node. The role of the backup node is to relocate to the failed node when the critical node fails, to avoid network partitioning due to node failure. It first follows the following principles to select a node from a one-hop neighbor as its backup node. The reason for selecting from a one-hop neighbor will be explained later:

[0037] Principle 1: The backup node cannot be a two-hop critical point and cannot be an RN node. This principle is to select a common node in the network that can be used for controllable movement.

[0038] Principle 2: If there are multiple nodes that meet Principle 1, select the node with a smaller node degree, because moving this node will have the least impact on other nodes;

[0039] Principle 3: If there are multiple nodes that meet Principle 2, in order to reduce energy consumption and shorten recovery time, choose a node that is closer to itself.

[0040] The information maintained by each actor is shown in Table 1. In addition to basic information such as node ID, location, critical status, and node type, each actor also maintains key information required for backup. The master node refers to the actor that uses the current node as a backup node. In the backup plan, a node may be selected as a backup for multiple actors. If the backup relationship is successfully established, when a node in the master node set loses its heartbeat, the backup node will move to the location of the master node. A special case is that the current node is an RN node but is selected as a backup node. This situation will be effectively handled in the subsequent backup and recovery process. In addition, the actor also maintains the length of the current recovery path.

[0041] Table 1 Node maintenance information

[0042] ID Location Critical state Backup node number Master Node Set Node Type Recovery costs 0 (x, y) YES / NO ID(node) Set(node) AN / CN / RN L

[0043] like Figure 3 This is a simple human-machine joint detection network topology consisting of four CN nodes and four RN nodes. After the network is initialized, CN3, RN2, and CN4 will be determined as two-hop critical points. After the determination is successful, these nodes will try to select a backup node from their one-hop neighbors according to the above principle. CN3 will select the closest node from CN1 and CN2 as the backup node. RN2 and CN4 cannot select a backup node according to the above principle because the neighbors of RN2 and CN4 are both two-hop critical points or RN nodes.

[0044] In order to back up all two-hop critical points in the network, it is not difficult to see that CN1 or CN2 can be used as the backup node of RN2. The problem at this time is that if the number of hops between CN1 and RN2 is large, resulting in a long repair distance between CN1 and RN2, if CN1 is moved directly to the location of RN2, it may cause CN1 to run out of energy and cause recovery failure. In order to improve the robustness of the network recovery mechanism, the backup node cannot be selected from a node outside one hop.

[0045] In order to ensure that every two-hop critical point can be backed up, for two-hop critical points that cannot select backup nodes according to the above principles, the type and critical status of the node will be ignored when selecting the backup. Every time a Hello message is received from a neighbor, the node that sends the Hello message will be checked to see if there is a backup node. If there is a backup node, the node that sends the Hello message will be regarded as its own backup. Therefore, after the network is initialized, CN3 will first regard CN1 as its own backup node. RN2 and CN4 cannot select a backup node. When CN1 and CN3 reach a consensus, the Hello message sent to RN2 will carry the information that CN3 has backed up. After receiving the Hello message, RN2 will choose CN3 as its own backup. Similar to the above method, CN4 will use RN2 as its temporary backup, and will offload the mobile task of the RN node during subsequent backup recovery. Figure 4 When RN2 fails, CN3 detects the loss of RN2's heartbeat message and moves to RN2's location. The movement will cause the link CN1-CN3 to fail, and CN1 will think that the CN3 node is faulty, so it moves to CN3's location. This cascade movement repair method divides a long moving distance task into several shorter moving tasks. Although the total moving distance may increase, it will not cause excessive movement of a single actor.

[0046] The selection of backups at the two-hop critical point is a local behavior, and the backup node is not aware of it. Therefore, when the two-hop critical point determines the backup, it will send a Commission message to the backup neighbor. The node that receives the Commission message will add the two-hop critical node that sent the Commission message to its own master node set and start monitoring the heartbeat. Because an actor can serve as a backup for multiple two-hop critical nodes, the master node also needs to monitor whether its own backup is alive. If its own backup fails or performs other backup tasks, the node will reselect a suitable backup from its neighbors.

[0047] The above backup mechanism is equivalent to generating a recovery chain when a failure occurs. The tail of the recovery chain must be a common controllable mobile node, and the middle node must be a two-hop critical node. If all nodes in the network are controllable and mobile, then a recovery chain can be found in the network for each critical point. However, for networks with RN nodes, because the above backup mechanism does not check the node type in special cases, there may be RN nodes in the recovery chain, such as Figure 3 The recovery chain of CN4-RN2-CN3-CN1 in the figure. CN4 regards RN2 as its own backup node. Considering the particularity of the detection network, although CN4 regards RN2 as a backup node, it will use RN2's backup node instead during actual recovery. Figure 5When RN2 detects that CN4's heartbeat is lost, it will send a Mission message to its backup node CN3, which contains CN4's location information. After receiving the Mission message, CN3 checks whether it is an RN node. If it is, it continues to forward the message to its backup. If not, it executes the mobile task. The above process is equivalent to adding a fault transmission mechanism during the recovery process, and offloading the recovery task of the RN node to the CN node with the smallest number of hops from itself on the recovery chain.

[0048] If a node has a backup node, the backup node path cost needs to be maintained , Here it is simplified to the total distance moved by all nodes to repair a faulty node. After receiving a Hello message with a lower recovery cost, the node will send a Cancel message to its own backup to cancel the backup plan and select a new backup node. Figure 6 A special network topology is given in which redundant nodes (circular nodes) are located at the edge of the network. Figure 6 The block node in (a) is the critical point that needs to be backed up after network initialization. Figure 6 The triangular nodes in (b) are nodes that have been successfully backed up by one-hop neighbors of all non-critical points in the network using local information. Other nodes cannot select backups from their own neighbors, so they will wait for their neighbors to successfully back up. Figure 6 (c) and Figure 6 As shown in (d) in the figure, when these two-hop critical nodes sense that a neighbor has successfully backed up, they will select the node with the lowest recovery cost from all optional backup nodes as the backup node. In this way, the backup information will continue to be transmitted to the unbacked nodes until all two-hop critical nodes are successfully backed up. The backup process of each node can be regarded as a tree structure. The root node of the tree is the node that needs to be backed up in the end. The backup plan starts from the leaf node and is transmitted layer by layer until the root node finally selects a backup path with the lowest recovery cost.

[0049] 3 Simulation Results and Analysis

[0050] The experiment first verifies the feasibility of the node backup mechanism. The experiment simulates networks with different node scales and different node type ratios. Each experimental result is the average value of 100 random initial connectivity graphs. The master control process uses OvS to create link connections between docker nodes. Because the more CN nodes in the network means the stronger the network reconstruction capability, the ratio of the number of CN nodes to the total number of nodes is defined as the network flexibility in this experiment. . Figure 7 (a) is constant The result of changing the network node scale when is 0.5, Figure 7 (b) in the figure is the result of changing the network flexibility with a constant network size of 60 nodes. The test results show that the success rate of cut point backup can reach 100% in different network scenarios. This is because the network backup mechanism can transmit the backup plan from the leaf node to each key node through the Hello message. Therefore, as long as there are redundant nodes in the network, the backup plan can be determined for each cut point. However, the tail of the recovery path of each cut point will gradually aggregate as the number of redundant nodes decreases.

[0051] When each node determines and changes the backup plan, it will output to the same log file. The experiment uses the difference between the last backup change time and the start time of the experiment to simply estimate the convergence time of the backup mechanism. Assume that the total number of nodes in the network is N , the communication radius is R , the network is deployed in The average degree of the adjacent nodes in the network is:

[0052] ,

[0053] It reflects the density of the entire network. The larger it is, the denser the network becomes. Figure 8 (a) is the fixed network flexibility is 0.5, and the average backup convergence time under different network scales and different adjacency characteristics is measured. From the results, it can be seen that as the network scale increases, no matter For which value, the average backup convergence time increases gradually. This is because as the network scale increases, the number of hops from the redundant CN nodes at the edge of the network to the network center node increases. The backup plan is transmitted through a scheduled Hello message, so when the number of hops of a backup plan path increases, its backup convergence time will also increase gradually. As it increases, the average convergence time of node backup will be shorter. Similar to the above principle, when the network is denser, the average shortest number of hops of each final backup path will also decrease, so the average backup convergence time will gradually decrease.

[0054] In order to verify the average backup convergence time under different network flexibility, the experiment measured N =80, the average backup convergence time is measured by changing the network flexibility and average adjacency. Figure 8As can be seen in (b), as the network flexibility gradually increases, the average backup convergence time will gradually decrease. This is because when the network flexibility increases, the number of non-critical CN nodes in the network will gradually increase, which will reduce the average backup hop count and the number of critical nodes that need to be backed up, thereby reducing the average backup convergence time. In any network state, the denser the network, the smaller the average backup convergence time.

[0055] The experiment also randomly verified the backup and recovery actions after the cut point failed under different network conditions. The results showed that when the cut point failed, the backup and recovery mechanism could relocate the node according to the backup and recovery chain without moving the RN node, thereby reconnecting the network.

[0056] It can be seen from the above simulation experiments that the node backup method of the present invention provides a feasible node failure recovery solution for the lunar man-machine joint exploration network. The mechanism makes full use of the redundant nodes in the network and allows the network to recover from the partition caused by the failure as much as possible without affecting the exploration mission of each astronaut. When the network is initialized, a backup node is selected for each key node. When the key node fails, a shortest backup recovery chain is formed to reconnect the network. The experiment verifies that the backup mechanism converges when there are redundant nodes in the network, and the success rate of the cut point backup can reach 100%.

Claims

1. A node backup method for a lunar man-machine joint exploration network, characterized in that: The method comprises the following steps: S1. Construct a lunar surface human-machine joint exploration network model: define the lunar surface gateway as an anchor point, each astronaut node as a random mobile node, and the mobile device in the exploration mission as a controllable mobile node, form a network in a wireless multi-hop manner, and obtain the coordinates of the anchor point, random mobile node, and controllable mobile node in the network; S2, in random mobile nodes and controllable mobile nodes, the k-hop cut point judgment algorithm is used to determine the two-hop critical point; S3, after the node is determined to be a two-hop critical point, it selects a node from its one-hop neighbor as its backup node; S4. When the two-hop critical node determines the backup, it will send a Commission message to its backup neighbor. The node that receives the Commission message will add the two-hop critical node that sent the Commission message to its own master node set and start monitoring the heartbeat. At the same time, the two-hop critical node that sends the Commission message also monitors whether its own backup is alive. If its own backup fails or performs other backup tasks, the two-hop critical node that sends the Commission message will return to step S3 to reselect the backup. S5. The two-hop critical point with backup needs to maintain the backup node path cost L, where L is the total distance moved by all nodes to repair a faulty node. When receiving a Hello message with a smaller recovery cost, the node will send a Cancel message to its own backup node to cancel the backup plan and select a new backup node. The selection principle of selecting a node from its one-hop neighbor as its backup node in step S3 is as follows: Principle 1: The backup node cannot be a two-hop critical point and cannot be a randomly moved node; Principle 2: If there are multiple nodes that meet Principle 1, select the node with the smallest node degree, because moving this node will have the least impact on other nodes; Principle 3: If there are multiple nodes that meet Principle 2, select the node closest to itself; For two-hop critical points that do not meet the principles one, two, and three, the type and critical status of the neighboring nodes of the two-hop critical points will be ignored when selecting backups. Each time a Hello message is received from a neighboring node, a check is made to see whether the neighboring nodes of the two-hop critical points have backup nodes. If there are backup nodes, the node that sends the hello message will be regarded as a backup of the two-hop critical point.

2. The node backup method for the lunar surface human-machine joint detection network according to claim 1 is characterized in that: The anchor points, random mobile nodes, and controllable mobile nodes in the network described in step S1 are all equipped with omnidirectional antennas, and the maximum transmission radius of each antenna is R max , when the distance between two nodes is less than R max A two-way communication link is considered to be established between them.

3. The node backup method for the lunar surface human-machine joint detection network according to claim 1 is characterized in that: The k-hop cut point judgment algorithm described in step S2 is used to determine the two-hop critical point. Specifically, if the node is a cut point of its own k-hop neighbor subgraph, the node is defined as a k-hop critical point. After the network is initialized, each node periodically sends a Hello broadcast message with a TTL of 2 to its surrounding neighbors, which carries basic information including its own ID, location, and node type so that other nodes can generate a two-hop subgraph to determine whether it is a two-hop critical point.

Images