A topology discovery method of whole-network dynamic perception

Abstract

The application discloses a kind of topology discovery methods of whole network dynamic perception, comprising the following steps: initialization whole network node information;Initialization node neighbor relation matrix;Carry out neighbor relation discovery;The matrix is split into master-slave area by the diagonal line of neighbor relation matrix;From area copy master area data, update the node that changes occurs in slave area, and switch master-slave area;From area copy master area data, add new node information and its neighbor relation information, switch master-slave area;Create delay queue, update new or changed node information;The timeout node in delay queue is re-performed neighbor relation discovery and is updated to registration center;Registration center and each node send heartbeat synchronization global neighbor relation information.The method described in the application reduces the complexity of transmission between nodes, alleviates the phenomenon of excessive resource consumption and heavy load, greatly improves the efficiency of topology discovery between nodes, and optimizes the network topology discovery method.

Description

Technical Field

[0001] This invention relates to the field of network device communication and management technology, and specifically to a method for dynamic topology discovery with full network awareness. Background Technology

[0002] With the increasing demands for interconnectivity and real-time performance of network resources in the "Internet+" era, network scale and structure have become increasingly large and complex. Therefore, an effective method for managing and maintaining network nodes is needed. Network topology discovery is a prerequisite for network device node management and maintenance. Currently, some traditional topology discovery methods mainly utilize link-layer discovery protocols to achieve network status awareness, but this results in significant resource consumption and increased controller load, leading to problems such as high load, complex information synchronization logic, and low efficiency. Therefore, how to reduce consumption and load and improve the efficiency of topology discovery is a pressing technical problem to be solved in this field. Summary of the Invention

[0003] The purpose of this invention is to propose a network-wide dynamic topology discovery method to solve the aforementioned technical problems. This invention is achieved through the following technical solution:

[0004] This invention proposes a topology discovery method based on dynamic sensing across the entire network, the method comprising the following steps:

[0005] S1: Initialize all network node information;

[0006] S2: Initialize the node neighbor relationship matrix;

[0007] S3: Perform neighbor relationship discovery and set the connectivity and bandwidth characteristics between nodes and their neighboring nodes;

[0008] S4: Divide the neighbor relationship matrix into a primary region and a secondary region by using the diagonal of the neighbor relationship matrix;

[0009] S5: Copy data from the master region to the slave region, update the nodes in the slave region where the neighbor relationships have changed, and switch between master and slave regions;

[0010] S6: Copy the data from the master region to the slave region, add information about the new node and its neighbor relationship information, update the neighbor relationship information of its neighbor nodes, and switch the master and slave regions;

[0011] S7: Create a delayed queue, add new or modified node information to the tail of the queue, and delete duplicate node information at the front of the queue.

[0012] S8: Rediscover the neighbor relationships of timed-out nodes in the delay queue. If there are any changes, update the registry center of the neighbor relationship matrix.

[0013] S9: The registry center sends heartbeats to each node to synchronize global neighbor relationship information.

[0014] Furthermore, the node information described in S1 includes node ID, node status, neighbor relationship confirmation status, and deployment location.

[0015] Furthermore, the initialization of the node neighbor relation matrix described in S2 specifically involves: retrieving the node relation expression C(a) from the node neighbor relation matrix Matrix(C). ij ,b ij The value assigned is: C(a) ij ,b ij ) = 0, where a ij For connectivity between nodes, b ij This represents the bandwidth characteristics between nodes.

[0016] Furthermore, the neighbor relationship discovery described in S3 includes the following steps:

[0017] S31: Nodes in the network use a MAC contention mechanism to elect a node. i () Send a hello message to the neighboring node;

[0018] S32: If the current node Node i If the state of a node conflicts with the states of some neighboring nodes, a backoff mechanism is used until a node that accepts the state is selected; if no conflict occurs, the node... i If all hello messages of () have been successfully sent and feedback has been received, then step S31 is executed repeatedly until a node that has not been successfully sent is selected to send a hello message.

[0019] S33: Neighbor nodes receive the hello message and extract the information, storing it in their own neighbor information set Set(Neighbor). i Simultaneously, feedback is provided within the process;

[0020] S34: Set(Neighbor) i The neighboring nodes that provide feedback transmit feedback signals in OFDM mode, including W subcarriers numbered sequentially from 1 to W; when node Node x When providing feedback, at the Zth... x Feedback is provided on each feedback bit, that is, a positive integer Z is randomly selected. x As a temporary designation, Z x ∈[1,W], the transmitted complex amplitude is S x Feedback signal:

[0021] S x =α / H x [Z x ]

[0022] Where α is a constant, H x [Z x Node i () to Node x () Inter-channel Z-th x The response on each subcarrier;

[0023] S35: Node i () The feedback result is sent to the neighboring node by sending a Response-Feedback message. After receiving the message, the neighboring node determines whether it has successfully fed back the feedback based on its own address. If it has successfully fed back the feedback, the field is set to 1; otherwise, it is set to 0.

[0024] S36: When the feedback of a node is confirmed, sufficient resources are reserved for the node that successfully fed back its feedback to send its ID+LOC+k+m in sequence, where k is the number of the neighbor node, k∈[1,m], and m is the number of the node that successfully fed back its feedback. The neighbor node receives the ID+LOC and stores it in its own neighbor information set.

[0025] S37: Repeat steps S31 to S36 until all neighboring nodes have been discovered, and then output the neighbor relationship matrix.

[0026] Furthermore, the diagonal of the neighbor relationship matrix described in S4 is formed by the node relationship expression C(a) of the neighbor relationship matrix Matrix(C). ij ,b ij The dividing line formed by the nodes i=j in ().

[0027] Furthermore, the copying of main region data from the region described in S5 and S6 specifically refers to, for the node relationship expression C(a ij ,b ij ), through operator C(a ij ,b ij )=C(a ji ,b ji Assigning values ​​enables data synchronization between master and slave regions.

[0028] Furthermore, the heartbeat sending described in S9 specifically involves the main area service node of the neighbor relationship matrix periodically sending requests to each individual node to synchronize the latest node neighbor relationships, thereby achieving global awareness.

[0029] The beneficial effects of this invention are as follows: The method described in this invention adopts a neighbor relationship matrix master-slave region switching and a delayed queue to update the node neighbor relationship, thereby obtaining the latest global neighbor relationship information. It then uses the heartbeat mechanism of the registration center to synchronize the latest neighbor relationship information to all nodes in the network in real time, transforming the traditional link discovery mechanism into a centralized proxy discovery mechanism. This allows for real-time dynamic perception of the global node relationships and status, avoiding the complexity of sequential synchronization between neighbor nodes in traditional topology discovery methods, and alleviating excessive resource consumption and heavy load. This significantly improves the efficiency of topology discovery between nodes, optimizing the network topology discovery method. It is suitable for large-scale and complex network application scenarios and has strong practicality.

[0030] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the method flow described in this invention;

[0032] Figure 2 A diagram illustrating the process of discovering neighbor relationships. Detailed Implementation

[0033] This invention proposes a topology discovery method based on dynamic sensing across the entire network, such as... Figure 1 As shown, the method includes the following steps:

[0034] S1: Initialize the node information of the entire network node set Set(Nodes). The node information includes node ID, node status (0 for sending, 1 for receiving), neighbor relationship confirmation status, deployment location, etc.

[0035] S2: Initialize the node neighbor relationship matrix Matrix(C), and define the node relationship expression C(a) of the node neighbor relationship matrix Matrix(C). ij ,b ij The value assigned is: C(a) ij ,b ij ) = 0, where a ij For connectivity between nodes, b ij This represents the bandwidth characteristics between nodes. Since the relationships between nodes are currently unknown, we assign a value of C(a). ij ,b ij If ) = 0, it means that all nodes are isolated and not connected to other nodes.

[0036] S3: Perform neighbor discovery, setting connectivity and bandwidth characteristics between the node and its neighbors. Neighbor discovery involves a node periodically sending request messages to attempt to establish a neighbor relationship with directly connected nodes. When a node receives a response message from the other node containing its own identifier, the neighbor relationship between the two nodes is successfully established. The sending period is typically 1 second. Specifically, for example... Figure 2 As shown, neighbor relationship discovery includes the following steps:

[0037] S31: Nodes in the network use a MAC contention mechanism to elect a node. i () Send a hello message to the neighboring node;

[0038] S32: If the current node Node i If the state of a node conflicts with the states of some neighboring nodes, a backoff mechanism is used until a node that accepts the state is selected; if no conflict occurs, the node... i If all hello messages of () have been successfully sent and feedback has been received, then step S31 is executed repeatedly until a node that has not been successfully sent is selected to send a hello message.

[0039] S33: Neighbor nodes receive the hello message and extract the information, storing it in their own neighbor information set Set(Neighbor). i Simultaneously, feedback is provided within the process;

[0040] S34: Set(Neighbor) i The neighboring nodes that provide feedback transmit feedback signals in OFDM mode, including W subcarriers numbered sequentially from 1 to W; when node Node x When providing feedback, at the Zth... x Feedback is provided on each feedback bit, that is, a positive integer Z is randomly selected. x As a temporary designation, Z x ∈[1,W], the transmitted complex amplitude is S x Feedback signal:

[0041] S x =α / H x [Z x ]

[0042] Where α is a constant, H x [Z x Node i () to Node x () Inter-channel Z-th x The response on each subcarrier;

[0043] S35: Node i(The feedback result is informed to the neighbor nodes by sending a Response-Feedback message. After receiving the message, the neighbor nodes judge whether their feedback is successful according to their own addressing. If the feedback is successful, the feedback field is set to 1, otherwise 0.)

[0044] S36: When the feedback of a node is confirmed, reserve sufficient resources for the nodes with successful feedback to send their ID + LOC + k + m in turn, where k is the number of the neighbor node, k ∈ [1, m], and m is the number of nodes with successful feedback. The neighbor nodes receive ID + LOC and store it in their neighbor information set.)

[0045] S37: Loop through steps S31 to S36 until all the nodes in the Set(Nodes) set are traversed, that is, all neighbor nodes are discovered, and then output the neighbor relationship matrix Matrix(C).

[0046] S4: Divide the neighbor relationship matrix into a main region and a slave region through the diagonal of the neighbor relationship matrix Matrix(C); the diagonal is a dividing line composed of the nodes where i = j in the node relationship expression C(a ij ,b ij ) of the neighbor relationship matrix Matrix(C), that is, the main and slave regions are divided into two regions according to C(a ij ,b ij ), i > j and i < j.)

[0047] S5: The slave region copies the data of the main region, updates the nodes with changed neighbor relationships in the slave region, and switches the main and slave regions. Specifically, for the node relationship expression C(a ij ,b ij ), the data synchronization between the main and slave regions is achieved by the operator C(a ij ,b ij ) = C(a ji ,b ji ).

[0048] S6: The slave region copies the data of the main region, and the copying method is the same as S5. Add the information of the newly added nodes and their neighbor relationships, and update the neighbor relationships of their neighbor nodes, and switch the main and slave regions.)

[0049] S7: Create a delay queue, add the information of the newly added or modified nodes to the end of the queue, and delete the repeated node information in front of the end of the queue. The delay queue updates the neighbor node information, that is, there is a delay task that scans the neighbor information set Set(Neighbor i ) of the nodes every 500ms, sends requests to the neighbor nodes in the set, waits for responses, and judges the connectivity a ij and the bandwidth characteristics bij Assigned to C(a) ij ,b ij ).

[0050] S8: Rediscover the neighbor relationships of timed-out nodes in the delay queue. If there are any changes, update the registry of the neighbor relationship matrix.

[0051] S9: The registration center sends heartbeats to each node to synchronize global neighbor relationship information. Specifically, sending heartbeats involves the main area service node of the neighbor relationship matrix periodically sending requests to each individual node to synchronize the latest node neighbor relationships, thereby achieving global awareness.

[0052] The topology discovery method described in this invention aims to achieve high-efficiency and low-load dynamic perception of all network nodes while meeting the communication requirements of complex network structures. Based on a registration center with a neighbor relationship matrix, it updates neighbor node information through a master-slave region CopyOnWrite mechanism. The registration center uses a heartbeat mechanism to synchronize the latest neighbor relationship information to all nodes in the network, thereby optimizing the node topology discovery scheme.

[0053] Finally, it should be noted that the above description is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred arrangement, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A topology discovery method with dynamic sensing across the entire network, characterized in that, The method includes the following steps: S1: Initialize all network node information; S2: Initialize the node neighbor relationship matrix; S3: Perform neighbor relationship discovery and set the connectivity and bandwidth characteristics between nodes and their neighboring nodes; The neighbor relationship discovery includes the following steps: S31: Nodes in the network use a MAC contention mechanism to elect a node. i () Send a hello message to the neighboring node; S32: If the current node Node i If the state of a node conflicts with the states of some neighboring nodes, a backoff mechanism is used until a node that accepts the state is selected; if no conflict occurs, the node... i If all hello messages of () have been successfully sent and feedback has been received, then step S31 is executed repeatedly until a node that has not been successfully sent is selected to send a hello message. S33: Neighbor nodes receive the hello message and extract the information, storing it in their own neighbor information set Set(Neighbor). i Simultaneously, feedback is provided within the process; S34: Set(Neighbor) i The neighboring nodes that provide feedback transmit feedback signals in OFDM, consisting of W subcarriers numbered sequentially from 1 to W; when node Node x When providing feedback, at the Zth... x Feedback is provided on each feedback bit, that is, a positive integer Z is randomly selected. x As a temporary designation, Z x [1,W], with a transmitted complex amplitude of S x Feedback signal: S x =α / H x [Z x ] Where α is a constant, H x [Z x Node i () to Node x () Inter-channel Z-th x The response on each subcarrier; S35: Node i () The feedback result is sent to the neighboring node by sending a Response-Feedback message. After receiving the message, the neighboring node determines whether it has successfully fed back the feedback based on its own address. If it has successfully fed back the feedback, the field is set to 1; otherwise, it is set to 0. S36: After a node's feedback is confirmed, sufficient resources are reserved for the node that successfully responded to send its ID + LOC + k + m sequentially, where k is the ID of the neighboring node, k [1, m], where m is the number of nodes that successfully responded. The neighboring node receives the ID+LOC and stores it in its own neighbor information set; S37: Repeat steps S31 to S36 until all neighboring nodes have been discovered, and then output the neighbor relationship matrix; S4: Divide the neighbor relationship matrix into a primary region and a secondary region by using the diagonal of the neighbor relationship matrix; S5: Copy data from the master region to the slave region, update the nodes in the slave region where the neighbor relationships have changed, and switch between master and slave regions; S6: Copy the data from the master region to the slave region, add information about the new node and its neighbor relationship information, update the neighbor relationship information of its neighbor nodes, and switch the master and slave regions; S7: Create a delayed queue, add new or modified node information to the tail of the queue, and delete duplicate node information at the front of the queue. S8: Rediscover the neighbor relationships of timed-out nodes in the delay queue. If there are any changes, update the registry center of the neighbor relationship matrix. S9: The registry center sends heartbeats to each node to synchronize global neighbor relationship information.

2. The topology discovery method for dynamic sensing across the entire network according to claim 1, characterized in that, The node information described in S1 includes node ID, node status, neighbor relationship confirmation status, and deployment location.

3. The topology discovery method for dynamic sensing across the entire network according to claim 1, characterized in that, The initialization of the node neighbor relation matrix described in S2 specifically involves: defining the node relation expression C(a) of the node neighbor relation matrix Matrix(C). ij ,b ij The value assigned is: C(a) ij , b ij )=0, where a ij For connectivity between nodes, b ij This represents the bandwidth characteristics between nodes.

4. The topology discovery method for dynamic sensing across the entire network according to claim 1, characterized in that, The diagonal of the neighbor relationship matrix described in S4 is formed by the node relationship expression C(a) of the neighbor relationship matrix Matrix(C). ij , b ij The dividing line formed by the nodes i=j in the diagram.

5. The topology discovery method for dynamic sensing across the entire network according to claim 1, characterized in that, Specifically, copying the main region data from the region as described in S5 and S6 refers to, for the node relationship expression C(a ij , b ij ), through operator C(a ij , b ij )= C(a ji , b ji Assigning values ​​enables data synchronization between master and slave regions.

6. The topology discovery method for dynamic sensing across the entire network according to claim 1, characterized in that, The heartbeat sending described in S9 specifically involves the main area service node of the neighbor relationship matrix periodically sending requests to each individual node to synchronize the latest node neighbor relationships, thereby achieving global awareness.

Images