Node identification method and device in edge scene and electronic equipment

Abstract

The application discloses a kind of node identification method and electronic equipment under edge scene, comprising: the hardware equipment characteristic information of collection edge node is constructed feature tree model;Using hash algorithm, the characteristics of edge node are associated with the ID identification pre-generated by the edge node, and stored in cloud database;When needing to identify edge node, according to feature tree model, the matching degree between the edge node and other edge nodes is calculated using fuzzy matching algorithm respectively;The matching degree obtained is compared with the preset matching degree threshold value, if greater than matching degree threshold value, then the two edge nodes are the same node, according to the characteristics of the identified edge node, the associated characteristics of edge node in cloud database are updated, otherwise, a new edge node is created in cloud database.The application is changed when node hardware, through feature tree and matching algorithm, it is convenient to manage and identify the characteristics of node, ensure the identification accuracy of node.

Description

Technical Field

[0001] This invention belongs to the field of edge device technology, specifically relating to a method, device, and electronic device for managing node identification in edge scenarios. Background Technology

[0002] In heterogeneous edge computing scenarios, the industry typically uses hardware serial numbers as identifiers to identify devices. However, frequent device firmware updates can lead to duplicate serial numbers, and frequent hardware changes due to malfunctions and other issues create a complex and chaotic situation regarding edge device characteristics. When tracking edge nodes is required, the serial number-based device identification method becomes inaccurate, resulting in problems such as node record conflicts and device loss. Summary of the Invention

[0003] To address the above problems, this invention proposes a node identification and management method, device, and electronic device for edge scenarios. Devices are identified by IDs, and device invariance is ensured through a strong association between the ID and device features. When features change, a fuzzy matching algorithm accurately identifies the device, thus solving the pain points of duplicate identifications for different types of devices, frequent hardware changes, and difficulty in unified tracking and management in edge scenarios. The technical solution adopted by this invention to solve the above technical problems is as follows:

[0004] A node identification management method for edge scenarios includes the following steps:

[0005] S1, Collect hardware device feature information of each edge node and construct a feature tree model of the edge node;

[0006] S2 uses a hash algorithm to associate the characteristics of an edge node with its pre-generated ID identifier and stores them in a cloud database;

[0007] S3. When it is necessary to identify edge nodes, the feature tree of the edge node is established according to the method of step S1, and the matching degree between the edge node and other edge nodes is calculated by using the fuzzy matching algorithm.

[0008] S4. Compare the matching degree obtained in step S3 with the preset matching degree threshold. If it is greater than the matching degree threshold, the two edge nodes are the same node. Update the association features of the edge nodes in the cloud database according to the characteristics of the identified edge nodes. Otherwise, create a new edge node in the cloud database according to the method of step S2.

[0009] Step S1 includes the following steps:

[0010] S1.1, Collect hardware device characteristic information for each edge node;

[0011] S1.2, Construct a feature tree for the edge nodes based on the features collected in step S1.1;

[0012] S1.3 sets weights for the root and leaf nodes of each feature tree to obtain the feature tree model.

[0013] The hardware device features include CPU supplier, CPU frequency, CPU serial number, MAC address, network card serial number, network card logical name, memory serial number, memory frequency, and memory supplier. Among these, CPU supplier, CPU frequency, and CPU serial number are CPU features; MAC address, network card serial number, and network card logical name are network card features; and memory serial number, memory frequency, and memory supplier are memory features.

[0014] The ID identifier is used to identify edge nodes, and different edge nodes have different ID identifiers.

[0015] The step of updating the associated features of edge nodes in the cloud database based on the features of the identified edge nodes specifically refers to associating the hash value corresponding to the features of the identified edge nodes, i.e., the new features of the edge nodes, with the ID identifier of the corresponding edge node in the database, so as to replace the original association between the ID identifier of the edge node in the database and its feature hash value.

[0016] A node identification management device for edge scenarios, comprising:

[0017] Feature tree building module: used to build feature tree models of edge nodes based on the collected hardware device feature information of edge nodes;

[0018] Association module: Used to associate the characteristics of each edge node with the pre-generated ID identifier of that edge node using a hash algorithm, and store the association in a cloud database;

[0019] Matching degree calculation module: used to calculate the matching degree between the edge node to be identified and other edge nodes based on the feature tree model constructed by the feature tree building module and the fuzzy matching algorithm;

[0020] Data update module: It is used to compare the matching degree calculated by the matching degree calculation module with the matching degree threshold. When it is greater than the matching degree threshold, it is determined that the two edge nodes are the same node. The association relationship in the cloud database is updated based on the characteristics of the edge node to be identified. Otherwise, a new edge node is created directly in the cloud database.

[0021] An electronic device includes a processor and a memory, the memory storing a computer program that, when executed by the processor, implements the node identification management method in an edge scenario as described above.

[0022] The beneficial effects of this invention are:

[0023] The ID identifier solves the problem of identifying physical nodes with duplicate serial numbers, ensuring the uniqueness of the identifier across the entire network. Furthermore, when node hardware changes, feature trees and matching algorithms facilitate the management and identification of node features, ensuring the accuracy of node identification. This solves the problems of duplicate identifiers for different types of devices, frequent hardware changes, and difficulty in unified tracking, management, and identification in edge scenarios. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of an embodiment of the feature tree of the present invention.

[0026] Figure 2 This is a schematic diagram of an embodiment of the feature tree model of the present invention.

[0027] Figure 3 This refers to the hardware device characteristic information of edge node A.

[0028] Figure 4 This refers to the hardware device characteristic information of edge node B.

[0029] Figure 5 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] A node identification management method in edge scenarios, such as Figure 5 As shown, it includes the following steps:

[0032] S1. Collect hardware device feature information of edge nodes and construct feature tree models of edge nodes, including the following steps:

[0033] S1.1, Collect hardware device characteristic information of edge nodes;

[0034] The hardware device feature information includes CPU supplier, CPU frequency, CPU serial number, MAC address, network card serial number, network card logical name, memory serial number, memory frequency, and memory supplier. Among them, CPU supplier, CPU frequency, and CPU serial number are CPU features, MAC address, network card serial number, and network card logical name are network card features, and memory serial number, memory frequency, and memory supplier are memory features.

[0035] S1.2, Construct a feature tree for the edge nodes based on the feature information collected in step S1.1;

[0036] A feature tree is a tree-like structure composed of all feature values ​​used to calculate the matching degree. In a feature tree, each child node represents a feature point that makes up the final feature.

[0037] S1.3, set weights for the root and leaf nodes of each feature tree to obtain the feature tree model;

[0038] like Figure 1 As shown, the overall features of the edge nodes are used as the root node, and the network interface card (NIC) features, CPU features, and memory features are used as child nodes of the root node. Similarly, the CPU vendor, CPU frequency, and CPU serial number are used as child nodes of the CPU feature. Likewise, the MAC address, NIC serial number, and NIC logical name are child nodes of the NIC feature, and the memory serial number, memory frequency, and memory vendor are child nodes of the memory feature. Then, feature weights are assigned to each node in the feature tree. Feature weight refers to the weight a node in the feature tree occupies in the fuzzy matching algorithm. The feature weight of the root node is 1, and the sum of the feature weights of all its sibling nodes is 1.

[0039] Specifically, the feature tree models are all pre-configured and generated by the user, and the feature weights of each node in the tree are also pre-configured by the user. The user can set the value of the feature weights according to the actual situation. As another embodiment of this application, when building the feature tree model, at least two features can be selected from network card features, CPU features, and memory features as child nodes of the root node. Alternatively, at least one lower-order feature can be selected from the first-order child nodes, i.e., network card features, CPU features, or memory features, as a second-order child node. For example, CPU features and memory features can be selected as child nodes of the root node, and the CPU vendor can be selected as a child node of the CPU feature, and the memory vendor and memory serial number can be selected as child nodes of the memory feature to construct the feature tree model. In addition, as Figure 1 As shown, depending on the specific situation, each of the child nodes can also be established below the second-order child nodes, namely the CPU serial number, CPU supplier, memory supplier, network card serial number, etc. Of course, in order to facilitate the management and identification of the features of the edge nodes in the later stage, the feature tree model of the edge nodes remains consistent in this embodiment.

[0040] S2, Based on the feature tree model established in step S1, the features of the edge node are associated with the pre-generated ID identifier of the edge node using a hash algorithm and stored in the cloud database;

[0041] The ID identifier is randomly generated when the edge node is first started, and the ID identifier of each edge node is fixed during the life cycle of each edge node, while the ID identifiers of different edge nodes are different.

[0042] Specifically, when establishing relationships for each edge node, the features of the edge node are first substituted into each node of the model according to the specific structure of the feature tree model, thus forming the feature tree corresponding to that edge node. The content corresponding to each node of the feature tree is sorted and concatenated in a fixed order. Then, the resulting string is hashed using an algorithm such as MD5. Finally, the hash value is associated with the ID identifier of the edge node. The relationships are stored in a cloud database for unified management. To avoid recognition confusion, the concatenation order of the feature strings of all edge nodes is kept consistent.

[0043] S3. When it is necessary to identify a certain edge node, based on the feature tree model established in step S1, the matching degree between the edge node and other edge nodes is calculated using the fuzzy matching algorithm.

[0044] like Figure 2 As shown, the feature weights of the child nodes of edge node 1 for CPU features, network card features, and memory features are as follows: the feature weight of CPU features is 0.5, the feature weight of network card features is 0.3, and the feature weight of memory features is 0.2. CPU supplier, CPU frequency, and CPU serial number are the child nodes of CPU features, and their corresponding feature weights are shown in the brackets in the figure. CPU architecture and CPU production time are the child nodes of CPU supplier.

[0045] Figure 3 The image shows the hardware device characteristics of edge node A. Figure 4 The diagram shows the hardware device feature information of edge node B. When performing fuzzy matching on two devices, all leaf nodes are compared first. Weights are calculated upwards according to the hierarchy. The similarity of all parent nodes is equal to the similarity of all child nodes multiplied by the sum of the feature weights of the corresponding child nodes. Therefore, according to... Figure 2 Feature weights, Figure 3 and Figure 4 The formula for calculating the similarity between them is:

[0046] S=0.5*[0.3*(0.6*0+0.4*1)+0.3*0+0.4*1]+0.3*(0.2*1+0.7*1+0.1*

[0047] 1)+0.2(0.3*1+0.5*1+0.2*0);

[0048] The similarity between edge node A and edge node B is calculated to be 0.86, therefore the matching degree between them is 86%.

[0049] S4. Compare the matching degree obtained in step S3 with the preset matching degree threshold. If it is greater than the matching degree threshold, then the two edge nodes are the same node. Update the association features of the corresponding edge nodes in the cloud database according to the characteristics of the identified edge nodes. Otherwise, create a new edge node in the cloud database according to the method of step S2.

[0050] The step of updating the associated features of the corresponding edge node in the cloud database based on the features of the identified edge node specifically refers to associating the features of the identified edge node, i.e., the features of the new edge node, with the ID identifier of the corresponding edge node in the database, so as to replace the original association between the ID identifier of the edge node and its features in the cloud database.

[0051] This application constructs a feature tree to monitor various features of edge nodes in real time, uses a hash algorithm to associate node features with their unique ID identifiers, which facilitates the management of hardware information of edge nodes, and uses matching degree to identify changing features, thus solving the problems of identifying and tracing edge nodes when hardware serial numbers are duplicated and when hardware features change.

[0052] This application embodiment also provides a node identification management device in edge scenarios, including:

[0053] Feature model building module: used to build feature tree models of edge nodes based on the collected hardware device feature information of edge nodes;

[0054] Association Module: Used to associate the features of each edge node with the pre-generated ID identifier of that edge node based on the feature model building module, and store them in the cloud database.

[0055] Matching degree calculation module: used to calculate the matching degree between the edge node to be identified and other edge nodes based on the feature tree model constructed by the feature tree building module and the fuzzy matching algorithm;

[0056] Data update module: It is used to compare the matching degree calculated by the matching degree calculation module with the matching degree threshold. When it is greater than the matching degree threshold, it is determined that the two edge nodes are the same node. The association relationship in the cloud database is updated based on the characteristics of the edge node to be identified. Otherwise, a new edge node is created directly in the cloud database.

[0057] This application also provides an electronic device, which includes a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the node identification management method in the edge scenario as described above.

[0058] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the node identification management method in edge scenarios as described above. Specifically, the storage medium can be a general-purpose storage medium, such as a removable disk or hard disk. When the computer program on the storage medium is run, it can execute the node identification management method in edge scenarios as described above.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A node identification management method in edge scenarios, characterized in that, Includes the following steps: S1, Collect hardware device feature information of each edge node and construct a feature tree model of the edge node; S2, based on the feature tree model and a hash algorithm, associates the features of the edge node with the pre-generated ID of the edge node and stores them in the cloud database; S3. When it is necessary to identify edge nodes, based on the feature tree model established in step S1, the matching degree between the edge node and other edge nodes is calculated using the fuzzy matching algorithm. S4. Compare the matching degree obtained in step S3 with the preset matching degree threshold. If it is greater than the matching degree threshold, the two edge nodes are the same node. Update the association features of the edge nodes in the cloud database according to the characteristics of the identified edge nodes. Otherwise, create a new edge node in the cloud database according to the method of step S2. The step of updating the associated features of edge nodes in the cloud database based on the features of the identified edge nodes specifically refers to associating the features of the identified edge nodes, i.e., the new features of the edge nodes, with the ID identifier of the corresponding edge node in the cloud database, so as to replace the original association relationship between the ID identifier of the edge node and its features in the cloud database. In step S3, the matching degree between edge nodes is obtained based on the similarity of the parent nodes of the feature tree model. The similarity of the parent node is equal to the similarity of all child nodes multiplied by the sum of the feature weights of the corresponding child nodes.

2. The node identification management method in edge scenarios according to claim 1, characterized in that, Step S1 includes the following steps: S1.1, Collect hardware device characteristic information for each edge node; S1.2, Construct a feature tree for the edge nodes based on the features collected in step S1.1; S1.3 sets weights for the root and leaf nodes of each feature tree to obtain the feature tree model.

3. The node identification management method in edge scenarios according to claim 2, characterized in that, The hardware device feature information includes CPU supplier, CPU frequency, CPU serial number, MAC address, network card serial number, network card logical name, memory serial number, memory frequency, and memory supplier. Among them, CPU supplier, CPU frequency, and CPU serial number are CPU features, MAC address, network card serial number, and network card logical name are network card features, and memory serial number, memory frequency, and memory supplier are memory features.

4. The node identification management method in edge scenarios according to claim 1, characterized in that, The ID identifier is used to identify edge nodes, and different edge nodes have different ID identifiers.

5. A node identification management device for edge scenarios, characterized in that, include: Feature tree building module: used to build feature tree models of edge nodes based on the collected hardware device feature information of edge nodes; Association module: Used to associate the features of each edge node with the pre-generated ID identifier of that edge node based on the feature tree model and a hash algorithm, and store the association in a cloud database; Matching degree calculation module: used to calculate the matching degree between the edge node to be identified and other edge nodes based on the feature tree constructed by the feature tree building module and the fuzzy matching algorithm; The data update module compares the matching degree calculated by the matching degree calculation module with the matching degree threshold. When the matching degree is greater than the matching degree threshold, it determines that the two edge nodes are the same node and updates the association relationship in the cloud database based on the features of the edge node to be identified. Otherwise, it directly creates a new edge node in the cloud database. The update of the associated features in the cloud database based on the features of the edge node to be identified specifically refers to associating the features of the identified edge node, i.e., the features of the new edge node, with the ID identifier of the corresponding edge node in the cloud database, so as to replace the original association relationship between the ID identifier of the edge node and its features in the cloud database. The matching degree between edge nodes is obtained based on the similarity of the parent nodes of the feature tree model. The similarity of the parent node is equal to the similarity of all child nodes multiplied by the sum of the feature weights of the corresponding child nodes.

6. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing a computer program, which, when executed by the processor, implements the node identification management method in an edge scenario as described in any one of claims 1-4.

Images