Method and device for processing poor quality of Internet of Things

By building a knowledge graph for poor quality in IoT, automatically diagnose and deal with poor quality problems in IoT nodes, the problems of poor quality diagnosis and low processing efficiency of poor quality in IoT are solved, and fast and accurate fault handling and cost savings are achieved.

CN115065584BActive Publication Date: 2024-11-15CHINA TELECOM CORP LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210782838.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-05
Publication Date
2024-11-15
Estimated Expiration
2042-07-05

AI Technical Summary

Technical Problem

The diagnosis and processing of poor quality of the Internet of Things has problems such as long time, high professional requirements, and easy to misjudgment. The complex business process leads to low quality processing efficiency.

Method used

By constructing a knowledge graph of quality difference of the Internet of Things, using the relationship sub-graph of the quality difference index and the reasons for quality difference and the relationship sub-graph of the reason for quality difference and the response operation of quality difference, we will automatically diagnose the causes of quality difference of the Internet of Things nodes and generate corresponding responses Operation, reduce manual intervention.

Benefits of technology

It improves the efficiency and accuracy of the diagnosis and troubleshooting of poor quality of the Internet of Things, saves labor costs, assists or replaces front-line operation and maintenance personnel to complete fault processing, and improves the interpretability of poor quality processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115065584B_ABST
    Figure CN115065584B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method and device for processing poor quality of the Internet of Things, and to the technical field of the Internet of Things. Among them, the method for processing poor quality of the Internet of Things includes: obtaining the poor quality index of the Internet of Things node; retrieving a first subgraph according to the poor quality index of the Internet of Things node to determine the poor quality cause of the Internet of Things node, wherein the first subgraph is a relationship subgraph between the poor quality index and the poor quality cause in the poor quality knowledge graph of the Internet of Things; retrieving a second subgraph according to the poor quality cause of the Internet of Things node to determine the poor quality response operation of the Internet of Things node, wherein the second subgraph is a relationship subgraph between the poor quality cause and the poor quality response operation in the poor quality knowledge graph of the Internet of Things. Through the above method, the efficiency and accuracy of the diagnosis and troubleshooting of poor quality of the Internet of Things can be improved, and the operation and maintenance personnel can be assisted or even replaced to complete the rapid and accurate handling of faults.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the technical field of Internet of Things, and in particular to a method and device for processing poor quality of the Internet of Things. Background Art

[0002] In recent years, with the continuous expansion of the scale of IoT business, IoT has problems with multiple quality diagnosis links and difficult positioning. Among them, poor quality refers to poor quality, which can be generally understood as failure. The rapid growth of existing business has brought great pressure to operation and maintenance personnel, and an intelligent auxiliary tool is urgently needed to reduce repetitive operation and maintenance work. In addition, the business process of IoT is complex, involving multiple links such as wireless network, core network, terminal, and platform. It is difficult to troubleshoot and often requires coordination among multiple departments, and the efficiency of quality processing is low.

[0003] Among the related technologies, the diagnosis and processing of IoT poor quality are mainly based on manual methods, which have the problems of long IoT poor quality processing time, high professional level requirements for personnel, and easy misjudgment. Summary of the invention

[0004] In view of the above technical problems, the present invention proposes a method and device for processing poor quality of the Internet of Things.

[0005] According to a first aspect of the present disclosure, a method for processing poor quality of an Internet of Things is provided, comprising: obtaining a poor quality indicator of an Internet of Things node; retrieving a first subgraph according to the poor quality indicator of the Internet of Things node to determine the poor quality cause of the Internet of Things node, wherein the first subgraph is a relationship subgraph between poor quality indicators and poor quality causes in a poor quality knowledge graph of the Internet of Things; retrieving a second subgraph according to the poor quality cause of the Internet of Things node to determine a poor quality response operation of the Internet of Things node, wherein the second subgraph is a relationship subgraph between poor quality causes and poor quality response operations in the poor quality knowledge graph of the Internet of Things.

[0006] In some embodiments, retrieving a first subgraph according to the quality difference index of the Internet of Things node to determine the quality difference cause of the Internet of Things node includes: when there is a common quality difference cause corresponding to all quality difference indicators of the Internet of Things node in the first subgraph, determining the quality difference cause of the Internet of Things node according to the common quality difference cause.

[0007] In some embodiments, retrieving the first subgraph according to the poor quality index of the Internet of Things node to determine the poor quality cause of the Internet of Things node also includes: when there is no common poor quality cause corresponding to all the poor quality indicators of the Internet of Things node in the first subgraph, determining multiple candidate poor quality causes according to the poor quality causes corresponding to all the poor quality indicators of the Internet of Things node; determining a recommended score for each of the multiple candidate poor quality causes; and determining the poor quality cause of the Internet of Things node according to the recommended score of each of the multiple candidate poor quality causes.

[0008] In some embodiments, determining a recommended score for each of the multiple candidate poor quality reasons includes: determining an associated quality difference indicator for each candidate poor quality reason among all quality difference indicators of the Internet of Things node; determining a relationship weight between each candidate poor quality reason and its associated quality difference indicator; and determining a recommended score for each candidate poor quality reason based on the relationship weight between each candidate poor quality reason and its associated quality difference indicator.

[0009] In some embodiments, determining the relationship weight between each candidate poor quality cause and its associated poor quality indicator includes: when the associated poor quality indicator is directly associated with the candidate poor quality cause, determining the relationship weight between the candidate poor quality cause and the associated quality difference indicator based on the weight of the edge between the candidate poor quality cause and the associated quality difference indicator in the first subgraph.

[0010] In some embodiments, determining the relationship weight between each candidate poor quality cause and its associated poor quality indicator also includes: when the associated poor quality indicator is indirectly associated with the candidate poor quality cause through an intermediate poor quality indicator, determining the relationship weight between the candidate poor quality cause and the associated quality poor indicator based on the weight of the edge between the candidate poor quality cause and the intermediate quality poor indicator in the first subgraph, and the weight of the edge between the intermediate quality poor indicator and the associated quality poor indicator.

[0011] In some embodiments, obtaining the quality difference indicator of the Internet of Things node includes: receiving an Internet of Things quality difference processing request; and obtaining the quality difference indicator of the Internet of Things node from the Internet of Things quality difference processing request.

[0012] In some embodiments, the quality difference knowledge graph of the Internet of Things also includes: a relationship subgraph between the Internet of Things node type and the quality difference type, a relationship subgraph between the Internet of Things node type and the quality difference index, a relationship subgraph between the quality difference type and the quality difference index, and a relationship subgraph between the quality difference type and the quality difference cause.

[0013] In some embodiments, it also includes: generating a work order according to the quality reasons of the Internet of Things node and the quality response operations of the Internet of Things node; and sending the work order to the user terminal.

[0014] In some embodiments, it also includes: updating the quality difference knowledge graph of the Internet of Things according to the quality difference processing result data fed back by the user terminal.

[0015] In some embodiments, it also includes: constructing an ontology model of the quality difference knowledge graph of the Internet of Things; and constructing the quality difference knowledge graph of the Internet of Things based on the ontology model of the quality difference knowledge graph of the Internet of Things.

[0016] According to a second aspect of the present disclosure, a device for processing poor quality of an Internet of Things is provided, comprising: an acquisition module configured to acquire a poor quality index of an Internet of Things node; a first determination module configured to retrieve a first subgraph according to the poor quality index of the Internet of Things node to determine the poor quality cause of the Internet of Things node, wherein the first subgraph is a relationship subgraph between poor quality indexes and poor quality causes in a poor quality knowledge graph of the Internet of Things; a second determination module configured to retrieve a second subgraph according to the poor quality cause of the Internet of Things node to determine the poor quality response operation of the Internet of Things node, wherein the second subgraph is a relationship subgraph between the poor quality cause and poor quality response operation in the poor quality knowledge graph of the Internet of Things.

[0017] According to a third aspect of the present disclosure, there is provided an Internet of Things quality difference processing device, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the Internet of Things quality difference processing method described in any of the above embodiments based on instructions stored in the memory.

[0018] According to a fourth aspect of the present disclosure, a computer storable medium is provided, on which computer program instructions are stored, and when the instructions are executed by a processor, the method for processing Internet of Things quality differences described in any of the above embodiments is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0020] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0021] Figure 1 is a flow chart showing a method for processing poor quality in the Internet of Things according to some embodiments of the present disclosure;

[0022] Figure 2 is a flow chart illustrating determining the cause of poor quality of an IoT node according to some embodiments of the present disclosure;

[0023] Figure 3a is a schematic diagram of the principle of calculating the relationship weight between the quality difference index and the quality difference cause according to some embodiments of the present disclosure;

[0024] Figure 3b is a schematic diagram of the principle of calculating the relationship weight between the quality difference index and the quality difference cause according to some embodiments of the present disclosure;

[0025] Figure 4 is a flowchart showing a method for processing poor quality of the Internet of Things according to other embodiments of the present disclosure;

[0026] Figure 5is a structural schematic diagram showing an ontology model of a quality difference knowledge graph of the Internet of Things according to some embodiments of the present disclosure;

[0027] Figure 6 is a block diagram showing an IoT quality difference processing device according to some embodiments of the present disclosure;

[0028] Figure 7 is a block diagram showing an IoT quality difference processing device according to other embodiments of the present disclosure;

[0029] Figure 8 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0030] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0031] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0032] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0033] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0035] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0036] Figure 1 FIG. 1 is a flow chart showing a method for processing poor quality of the Internet of Things according to some embodiments of the present disclosure. Figure 1 As shown, the method for processing poor quality of the Internet of Things in the embodiment of the present disclosure includes:

[0037] Step S110: Obtain the quality index of the IoT node.

[0038] In some embodiments, an IoT poor quality processing request is received, and a poor quality index of an IoT node is obtained from the IoT poor quality processing request. For example, after a certain IoT node fails, it sends an IoT poor quality processing request to an IoT poor quality processing device, wherein the IoT poor quality processing request includes information such as an identifier of the IoT node and a poor quality index. The IoT poor quality processing device parses the IoT poor quality processing request to obtain the poor quality index of the IoT node.

[0039] In other embodiments, the IoT quality difference processing device monitors one or more IoT nodes regularly or periodically, and actively obtains the quality difference index of an IoT node when a failure of an IoT node is detected.

[0040] Exemplarily, the IoT node is a terminal side node, a wireless network side node, or a core network side node in the IoT.

[0041] In some embodiments, the quality index of the IoT node is an abnormal performance index of the IoT node. For example, when a water meter terminal of a water meter factory fails to connect to the network or allocate an IP address, the water meter terminal sends quality indexes such as a sharp increase in uplink and downlink throughput, a sharp increase in uplink and downlink subcarrier utilization, and a sharp increase in uplink signal interference to the IoT quality processing device.

[0042] Step S120: searching the first subgraph according to the poor quality index of the IoT node to determine the cause of the poor quality of the IoT node.

[0043] Among them, the first subgraph is the relationship subgraph between quality difference indicators and quality difference causes in the quality difference knowledge graph of the Internet of Things.

[0044] In step S120, the relationship subgraph between the poor quality index and the poor quality cause is retrieved according to the poor quality index of the IoT node, and the poor quality cause corresponding to the poor quality index of the IoT node can be obtained. Then, the poor quality cause of the IoT node is determined according to the poor quality cause corresponding to the poor quality index of the IoT node. For example, the poor quality index of the IoT node includes low service uplink success rate, low radio resource control (RRC) connection success rate, and low cell uplink subcarrier utilization rate. The poor quality cause corresponding to the poor quality index determined by retrieving the first subgraph is: individual terminals frequently send data packets to generate a large number of connections, and the wireless side base station cell parameter indicator configuration needs to be optimized. After determining these two poor quality causes, the poor quality cause of the IoT node is determined according to these two poor quality causes.

[0045] In some embodiments, the Internet of Things quality difference processing method also includes: constructing an ontology model of the quality difference knowledge graph of the Internet of Things; constructing the quality difference knowledge graph of the Internet of Things based on the ontology model of the quality difference knowledge graph of the Internet of Things, and then storing the knowledge graph in a graph database.

[0046] Step S130: searching the second subgraph according to the quality problem of the IoT node to determine the quality problem response operation of the IoT node.

[0047] Among them, the second subgraph is the relationship subgraph between the causes of poor quality and the operations to deal with poor quality in the poor quality knowledge graph of the Internet of Things.

[0048] In step S130, the relationship subgraph between the quality difference reason and the quality difference coping operation is searched according to the quality difference reason of the IoT node, and the quality difference coping operation corresponding to the quality difference reason of the IoT node is obtained. Then, the quality difference coping operation of the IoT node is determined according to the quality difference coping operation corresponding to the quality difference reason of the IoT node. For example, the quality difference coping operation corresponding to the quality difference reason of the IoT node is used as the quality difference coping operation of the IoT node.

[0049] In the disclosed embodiment, the above steps can realize the automatic diagnosis and processing of poor quality of the Internet of Things, improve the efficiency and accuracy of poor quality diagnosis and troubleshooting of the Internet of Things, effectively save labor costs, and assist or even replace front-line operation and maintenance personnel to complete the rapid and accurate processing of faults. In addition, compared with poor quality diagnosis and troubleshooting based on neural network models, the method of the disclosed embodiment is more explainable and conforms to the usage habits of front-line operation and maintenance personnel.

[0050] Figure 2 FIG. 1 is a flow chart illustrating determining the cause of poor quality of an IoT node according to some embodiments of the present disclosure. Figure 2 As shown, the process of determining the cause of poor quality of an IoT node in the embodiment of the present disclosure includes:

[0051] Step S121: Retrieve the first subgraph according to the quality difference index of the IoT node.

[0052] Among them, the first subgraph is a relationship subgraph between poor quality indicators and poor quality reasons in the poor quality knowledge graph of the Internet of Things. In some embodiments, after receiving the poor quality processing request of the Internet of Things, the first subgraph is retrieved according to the poor quality indicators of the Internet of Things node carried in the poor quality processing request of the Internet of Things, so as to determine the poor quality reasons corresponding to each poor quality indicator of the Internet of Things node.

[0053] Step S122: Determine whether there is a common quality difference reason.

[0054] In step S122, it is determined whether there is a common poor quality cause corresponding to all poor quality indicators of the IoT nodes in the first subgraph. If there is a common poor quality cause, step S123 is executed; if there is no common poor quality cause, step S124 is executed.

[0055] For example, Figure 3aAs shown, assuming that the quality difference indicators of the IoT node are a, b, and c, by searching the first subgraph, it is found that the quality difference indicators a, b, and c all correspond to the quality difference reason 2, thereby judging that there is a common quality difference reason, and then executing step S123.

[0056] For example, Figure 3b As shown, assuming that the quality difference indicators of the Internet of Things nodes are d, e, and f, by searching the first subgraph, it is found that the quality difference indicator d corresponds to the quality difference reason 1, the quality difference indicator e corresponds to 2, and the quality difference indicator f corresponds to the quality difference reason 2. It is thus determined that there is no common quality difference reason, and step S124 is subsequently executed.

[0057] Step S123: Determine the poor quality cause of the IoT node according to the common poor quality cause.

[0058] In some embodiments, a common poor quality cause corresponding to all poor quality indicators of the IoT node is used as the poor quality cause of the IoT node.

[0059] In other embodiments, when there is one common poor quality cause corresponding to all poor quality indicators of the Internet of Things node, the common poor quality cause corresponding to all poor quality indicators of the Internet of Things node is used as the poor quality cause of the Internet of Things node; when there are multiple common poor quality causes corresponding to all poor quality indicators of the Internet of Things node, the recommended score of each common poor quality cause is determined, and the common poor quality cause with the largest recommended score is used as the poor quality cause of the Internet of Things node.

[0060] Step S124: Determine a plurality of candidate poor quality reasons according to the poor quality reasons corresponding to all the poor quality indicators of the IoT node.

[0061] In some embodiments, the poor quality reasons corresponding to all the poor quality indicators of the IoT node are used as candidate poor quality reasons. Figure 3b As shown, assuming that the poor quality indicators of the IoT node are d, e, and f, by searching the first subgraph, it is found that the poor quality indicator d corresponds to the poor quality reason 1, the poor quality indicator e corresponds to the poor quality reason 2, and the poor quality indicator f corresponds to the poor quality reason 2, then the poor quality reasons 1 and 2 are taken as candidate poor quality reasons. Exemplarily, the poor quality indicator d is the low uplink success rate of the service, the poor quality indicator e is the low RRC connection success rate, the poor quality indicator f is the low uplink subcarrier utilization rate of the cell, the poor quality reason 1 is that individual terminals frequently send data packets to generate a large number of connections, and the poor quality reason 2 is that the parameter indicator configuration of the base station cell on the wireless side needs to be optimized.

[0062] In other embodiments, a poor quality cause that meets a preset condition is selected from the poor quality causes corresponding to all the poor quality indicators of the IoT node, and the selected poor quality cause is used as a candidate poor quality cause.

[0063] Step S125: Determine a recommendation score for each of the plurality of candidate poor quality reasons.

[0064] In some embodiments, step S125 includes:

[0065] Step S1251: Determine the associated poor quality indicator of each candidate poor quality cause among all poor quality indicators of the IoT node.

[0066] In some embodiments, the associated poor quality indicator of the candidate poor quality cause refers to the poor quality indicator included in the poor quality indicators of the Internet of Things node and connected to the candidate poor quality cause in the first subgraph.

[0067] For example, Figure 3b As shown, it is assumed that the quality indicators of the Internet of Things node are d, e, and f, and the candidate quality reasons are quality reasons 1 and 2. The associated quality indicator of quality reason 1 refers to the quality indicator included in the quality indicator of the Internet of Things node and connected to quality reason 1, that is, quality indicator d; the associated quality indicator of quality reason 2 refers to the quality indicator included in the quality indicator of the Internet of Things node and connected to quality reason 2, that is, quality indicators e and f. Exemplarily, quality indicator d is a low service uplink success rate, quality indicator e is a low RRC connection success rate, quality indicator f is a low cell uplink subcarrier utilization rate, quality reason 1 is that individual terminals frequently send data packets to generate a large number of connections, and quality reason 2 is that the wireless side base station cell parameter indicator configuration needs to be optimized.

[0068] Step S1252: Determine the relationship weight between each candidate poor quality reason and its associated poor quality indicator.

[0069] In some embodiments, when the associated poor quality indicator is directly associated with the candidate poor quality cause, the relationship weight between the candidate poor quality cause and the associated poor quality indicator is determined according to the weight of the edge between the candidate poor quality cause and the associated poor quality indicator in the first subgraph. For example, the weight of the edge between the candidate poor quality cause and the associated poor quality indicator in the first subgraph is used as the relationship weight between the candidate poor quality cause and the associated poor quality indicator.

[0070] For example, Figure 3b As shown in Figure 1, assume that the quality indicators of the IoT nodes are d, e, and f, and the candidate quality reasons are quality reasons 1 and 2. For quality reason 1, the associated quality indicator is d, and d is directly associated with quality reason 1. Therefore, the weight q of the edge between quality reason 1 and quality indicator d is 1-d , as the relationship weight between poor quality cause 1 and the associated poor quality index.

[0071] In some embodiments, when the associated poor quality indicator is indirectly associated with the candidate poor quality cause through the intermediate poor quality indicator, the relationship weight between the candidate poor quality cause and the associated poor quality indicator is determined according to the weight of the edge between the candidate poor quality cause and the intermediate poor quality indicator, and the weight of the edge between the intermediate poor quality indicator and the associated poor quality indicator in the first subgraph. For example, the weight of the edge between the candidate poor quality cause and the intermediate poor quality indicator, and the weight of the edge between the intermediate poor quality indicator and the associated poor quality indicator are multiplied, and the result of the operation is used as the relationship weight between the candidate poor quality cause and the associated poor quality indicator.

[0072] For example, Figure 3b As shown in Figure 1, assume that the quality indicators of the IoT nodes are d, e, and f, and the candidate quality reasons are quality reasons 1 and 2. For quality reason 2, the associated quality indicators are e and f. Among them, quality indicator e is indirectly connected to quality reason 2 through quality indicator f. Therefore, the weight q of the edge between quality reason 2 and quality indicator f is 2-f , and the weight q of the edge between the quality difference index f and the quality difference index e e-f The product of q 2-f *q e-f , as the relationship weight between poor quality reason 1 and poor quality indicator e. For example, poor quality indicator d is low service uplink success rate, poor quality indicator e is low RRC connection success rate, poor quality indicator f is low cell uplink subcarrier utilization, poor quality reason 1 is that individual terminals frequently send data packets to generate a large number of connections, and poor quality reason 2 is that the wireless side base station cell parameter indicator configuration needs to be optimized.

[0073] In the embodiment of the present disclosure, when calculating the recommended score of the cause of poor quality, not only the direct correlation between the poor quality indicator and the cause of poor quality is considered, but also the indirect correlation between the poor quality indicator and the cause of poor quality is considered, which helps to improve the calculation accuracy of the recommended score of the cause of poor quality and make the IoT poor quality diagnosis more accurate.

[0074] Step S1253: Determine the recommended score of each candidate poor quality reason according to the relationship weight between each candidate poor quality reason and its associated poor quality indicator.

[0075] In some embodiments, the cumulative sum of the relationship weights between a candidate poor quality cause and all associated poor quality indicators is used as the recommendation score of the candidate poor quality cause.

[0076] For example, Figure 3b As shown in Figure 1, it is assumed that the poor quality indicators of the IoT nodes are d, e, and f, and the candidate poor quality reasons are poor quality reasons 1 and 2. For poor quality reason 1, the associated poor quality indicator is poor quality indicator d, and the recommended score of poor quality reason 1 is equal to the relationship weight q between poor quality reason 1 and poor quality indicator d. 1-dFor the quality reason 2, the associated quality indicators are quality indicators e and f, and the recommendation score of the quality reason 2 is equal to the relationship weight q between the quality reason 2 and the quality indicator e. e-f *q 2-f , and the relationship weight q between the quality difference reason 2 and the quality difference index f 2-f The sum of e-f *q 2-f +q 2-f .

[0077] Step S126: Determine the poor quality reason of the IoT node according to the recommendation score of each of the multiple candidate poor quality reasons.

[0078] In some embodiments, the candidate poor quality cause with the largest recommendation score is used as the poor quality cause of the IoT node.

[0079] In the disclosed embodiment, the above steps can quickly and accurately determine the quality reasons of IoT nodes from the IoT quality knowledge graph, improve the efficiency and accuracy of IoT quality diagnosis and troubleshooting, effectively save labor costs, and assist or even replace front-line operation and maintenance personnel to complete the rapid and accurate handling of faults. In addition, compared with quality diagnosis and troubleshooting based on neural network models, the method of the disclosed embodiment is more explainable and conforms to the usage habits of front-line operation and maintenance personnel.

[0080] Figure 4 FIG. 1 is a flow chart showing a method for processing poor quality of the Internet of Things according to other embodiments of the present disclosure. Figure 4 As shown, the method for processing poor quality of the Internet of Things in the embodiment of the present disclosure includes:

[0081] Step S410: receiving an IoT poor quality processing request.

[0082] In some embodiments, after a failure occurs, the IoT node sends an IoT poor quality processing request to the IoT poor quality processing device, and the IoT poor quality processing device receives the IoT poor quality processing request.

[0083] Exemplarily, the IoT quality processing request includes the IoT node identifier, IoT node quality index and other data. For example, the IoT quality processing request sent by the terminal side node carries the node's service type, network access method, sector, district, quality index and other data; the IoT quality processing request sent by the wireless network side node carries the node's sector, base station identifier, district, province, quality index and other data; the IoT quality processing request sent by the core network side node carries the node's network element type, region, quality index and other data.

[0084] Step S420: Obtain the quality difference index of the IoT node from the IoT quality difference processing request.

[0085] In step S420, the IoT quality difference processing request is parsed to obtain the quality difference index of the IoT node.

[0086] Step S430: searching the first subgraph according to the poor quality index of the IoT node to determine the cause of the poor quality of the IoT node.

[0087] Among them, the first subgraph is the relationship subgraph between quality difference indicators and quality difference causes in the quality difference knowledge graph of the Internet of Things.

[0088] In step S430, the relationship subgraph between the poor quality index and the poor quality cause is retrieved according to the poor quality index of the Internet of Things node, and the poor quality cause corresponding to the poor quality index of the Internet of Things node can be obtained. Then, according to the poor quality cause corresponding to the poor quality index of the Internet of Things node, the poor quality cause of the Internet of Things node is determined.

[0089] In some embodiments, when there is a common poor quality cause corresponding to all poor quality indicators of the IoT node in the first subgraph, the poor quality cause of the IoT node is determined according to the common poor quality cause. For example, the common poor quality cause is used as the poor quality cause of the IoT node.

[0090] In some embodiments, when there is no common poor quality cause corresponding to all poor quality indicators of the IoT node in the first subgraph, multiple candidate poor quality causes are determined according to the poor quality causes corresponding to all poor quality indicators of the IoT node; and the poor quality cause of the IoT node is determined according to the recommended score of each of the multiple candidate poor quality causes. For example, the candidate poor quality cause with the largest recommended score is used as the poor quality cause of the IoT node.

[0091] Step S440: searching the second subgraph according to the quality problem of the IoT node to determine the quality problem response operation of the IoT node.

[0092] Among them, the second subgraph is a relationship subgraph between the quality difference cause and the quality difference response operation in the quality difference knowledge graph of the Internet of Things. In some embodiments, in addition to the first subgraph and the second subgraph, the quality difference knowledge graph of the Internet of Things also includes: a relationship subgraph between the Internet of Things node type and the quality difference type, a relationship subgraph between the Internet of Things node type and the quality difference index, a relationship subgraph between the quality difference type and the quality difference index, and a relationship subgraph between the quality difference type and the quality difference cause.

[0093] In step S440, the relationship subgraph between the quality difference reason and the quality difference coping operation is searched according to the quality difference reason of the IoT node, and the quality difference coping operation corresponding to the quality difference reason of the IoT node is obtained. Then, the quality difference coping operation of the IoT node is determined according to the quality difference coping operation corresponding to the quality difference reason of the IoT node. For example, the quality difference coping operation corresponding to the quality difference reason of the IoT node is used as the quality difference coping operation of the IoT node.

[0094] In some embodiments, the method for processing quality difference of the Internet of Things also includes: constructing an ontology model of a quality difference knowledge graph of the Internet of Things; and constructing a quality difference knowledge graph of the Internet of Things based on the ontology model of the quality difference knowledge graph of the Internet of Things.

[0095] In some embodiments, constructing an ontology model of the quality difference knowledge graph of the Internet of Things includes: collecting historical quality difference processing data of the Internet of Things, performing data cleaning on the historical quality difference processing data of the Internet of Things to obtain historical sample data; and constructing an ontology model of the quality difference knowledge graph of the Internet of Things based on the historical sample data and domain expert experience information. The historical quality difference processing data of the Internet of Things includes: the original text data of the user-reported fault processing results and user-reported fault descriptions of various types of products of the Internet of Things.

[0096] Among them, the ontology model of the quality difference knowledge graph of the Internet of Things includes multiple entities and the relationships between multiple entities. For example, the ontology model of the quality difference knowledge graph of the Internet of Things includes the entities of the Internet of Things node type, quality difference type, quality difference index, quality difference reason, and quality difference response operation, as well as the relationship between the Internet of Things node type and quality difference type, the relationship between the Internet of Things node type and quality difference index, the relationship between the quality difference type and quality difference index, the relationship between the quality difference type and quality difference reason, the relationship between the quality difference index and quality difference reason, and the relationship between the quality difference reason and quality difference response operation.

[0097] In some embodiments, a quality difference knowledge graph of the Internet of Things is constructed based on the ontology model of the quality difference knowledge graph of the Internet of Things, including: extracting knowledge from historical sample data obtained through data cleaning to obtain detailed entity relationships based on the ontology model of the quality difference knowledge graph of the Internet of Things; determining the weight of the relationship between entities, that is, the weight of the edge between entities, based on the historical sample data; and storing the relationship between entities and the weights of the relationship in a graph database.

[0098] Exemplarily, the weight of the edge between the poor quality cause and the poor quality response operation is determined based on the number of each poor quality response operation (or business maintenance operation) corresponding to the poor quality cause according to the historical sample data statistics, as well as the total number of all poor quality response operations corresponding to the poor quality cause. For example, assuming that poor quality cause 1 corresponds to poor quality response operations L1, L2, and L3, wherein poor quality response operation L1 is optimized for downlink resources, poor quality response operation L2 is optimized for the number of uplink retransmissions, and poor quality response operation L3 is optimized for the networking mode of the Internet of Things nodes. According to the historical sample data statistics, the number of poor quality response operations L1, L2, and L3 are 5, 10, and 5, respectively, and the total number of poor quality response operations corresponding to poor quality cause 1 is 20. Then, the number of poor quality response operations L1, L2, and L3 can be set to account for 0.25, 0.5, and 0.25, which are used as the weight of the edge between poor quality cause 1 and poor quality response operations L1, L2, and L3.

[0099] In addition, the number of poor quality reasons traced by the poor quality indicators and the number of poor quality categories of various IoT nodes can be counted according to the historical sample data, and the weight of the edge between the poor quality indicators and the poor quality reasons, the weight of the edge between the IoT node type and the poor quality type, etc. can be determined accordingly. In addition, the correlation between the poor quality indicators can be counted according to the historical sample data, and the weight of the edge between the poor quality indicators can be determined according to the correlation between the poor quality indicators. For example, the poor quality indicator is that the in-use terminal has no business data reported for many days. The poor quality reasons corresponding to the poor quality indicator include the following four categories: weak network coverage, no response to the network side connection, terminal configuration error, network authentication failure, and terminal failure. If the frequency of occurrence of the above four types of poor quality reasons is 2, 3, 3, and 12 according to the historical sample data, the weights of the edges between the poor quality indicator and these four types of poor quality reasons are 0.1, 0.15, 0.15, and 0.6, respectively.

[0100] In the disclosed embodiment, the construction of the IoT quality difference knowledge graph is realized through the three links of ontology modeling, knowledge extraction, and weight assignment. Deductive reasoning is performed on the quality difference phenomena in various links of the IoT, and the semantic primary and secondary relationships in the direction of fault diagnosis are effectively distinguished, making the constructed knowledge graph more accurate.

[0101] Step S450: Generate a work order according to the quality reasons of the IoT nodes and the operation to deal with the quality problems of the IoT nodes.

[0102] After obtaining the quality difference reasons and quality difference coping operations of the IoT node to be maintained by searching the quality difference knowledge graph of the IoT, a work order is generated. Exemplarily, the work order includes information such as the identification of the IoT node to be maintained, quality difference indicators, quality difference reasons, quality difference coping operations, etc.

[0103] Step S460: Send the work order to the user terminal.

[0104] In some embodiments, the work order is sent to the terminal device corresponding to the maintenance personnel to assist the maintenance personnel in troubleshooting. Through steps S450 to S460, the efficiency of IoT quality control can be further improved.

[0105] In some embodiments, the Internet of Things quality difference processing method further includes: updating the quality difference knowledge graph of the Internet of Things according to the quality difference processing result data fed back by the user terminal. For example, after the maintenance personnel perform troubleshooting according to the recommended quality difference response operation, they feed back the quality difference processing result data such as whether the recommended quality difference response operation is correct to the Internet of Things quality difference processing device, and the Internet of Things quality difference processing device optimizes and updates the quality difference knowledge graph of the Internet of Things according to the feedback information.

[0106] In the disclosed embodiment, the above steps realize automatic diagnosis and troubleshooting of IoT faults, improve the efficiency and accuracy of IoT poor quality diagnosis and troubleshooting, effectively save labor costs, and assist or even replace front-line operation and maintenance personnel to complete the rapid and accurate handling of faults. In addition, compared with poor quality diagnosis and troubleshooting based on neural network models, the method of the disclosed embodiment is more explainable and conforms to the usage habits of front-line operation and maintenance personnel.

[0107] Figure 5 is a schematic diagram showing the structure of an ontology model of a quality difference knowledge graph of the Internet of Things according to some embodiments of the present disclosure. Figure 5 As shown, the ontology model of the quality difference knowledge graph of the Internet of Things in the embodiment of the present disclosure includes seven types of entities and 11 types of relationships.

[0108] Among them, the seven types of entities include: terminal side entity, wireless network side entity, core network side entity, poor quality category entity, poor quality indicator entity, poor quality cause entity, and poor quality response operation entity. Among them, the terminal side entity, wireless network side entity and core network side entity are IoT node type entities. The 11 types of relationships include: the relationship between the terminal side entity and the poor quality type entity, the relationship between the wireless network side entity and the poor quality type entity, the relationship between the core network side entity and the poor quality type entity, the relationship between the terminal side entity and the poor quality indicator entity, the relationship between the wireless network side entity and the poor quality indicator, the relationship between the core network side entity and the poor quality indicator entity, the relationship between the poor quality category entity and the poor quality indicator entity, the relationship between the poor quality indicator entity, the relationship between the poor quality category entity and the poor quality cause entity, the relationship between the poor quality indicator entity and the poor quality cause entity, and the relationship between the poor quality cause entity and the poor quality response operation entity.

[0109] In some embodiments, according to Figure 5 The quality difference knowledge graph of the Internet of Things constructed by the ontology model shown includes the following subgraphs: the relationship subgraph between terminal-side entities and quality difference types, the relationship subgraph between wireless network-side entities and quality difference types, the relationship subgraph between core network-side entities and quality difference types, the relationship subgraph between terminal-side entities and quality difference indicators, the relationship subgraph between wireless network-side entities and quality difference indicators, the relationship subgraph between core network-side entities and quality difference indicators, the relationship subgraph between quality difference types and quality difference indicators, as well as the relationship subgraph between quality difference types and quality difference causes, the relationship subgraph between quality difference indicators and quality difference causes, and the relationship subgraph between quality difference causes and quality difference response operations.

[0110] Figure 6 is a block diagram showing an IoT quality difference processing device according to some embodiments of the present disclosure. Figure 6 As shown, the IoT quality difference processing device 600 of the embodiment of the present disclosure includes: an acquisition module 610, a first determination module 620, and a second determination module 630.

[0111] The acquisition module 610 is configured to acquire the quality difference index of the IoT node.

[0112] In some embodiments, the acquisition module 610 receives an IoT quality difference processing request, and acquires a quality difference indicator of an IoT node from the IoT quality difference processing request.

[0113] In other embodiments, the acquisition module 610 monitors one or more IoT nodes regularly or periodically, and actively acquires the quality index of an IoT node when a failure of an IoT node is detected.

[0114] Exemplarily, the IoT node is a terminal side node, a wireless network side node, or a core network side node in the IoT.

[0115] In some embodiments, the quality index of the IoT node is an abnormal performance index of the IoT node. For example, when a water meter terminal of a water meter factory fails to connect to the network or allocate an IP address, the water meter terminal of the water meter factory sends quality indexes such as a sharp increase in uplink and downlink throughput, a sharp increase in uplink and downlink subcarrier utilization, and a sharp increase in uplink signal interference to the IoT quality processing device.

[0116] The first determination module 620 is configured to retrieve the first subgraph according to the poor quality index of the IoT node to determine the cause of the poor quality of the IoT node.

[0117] Among them, the first subgraph is the relationship subgraph between quality difference indicators and quality difference causes in the quality difference knowledge graph of the Internet of Things.

[0118] The first determination module 620 retrieves the relationship subgraph between the quality difference index and the quality difference cause according to the quality difference index of the Internet of Things node, and obtains the quality difference cause corresponding to the quality difference index of the Internet of Things node. Then, the first determination module 620 determines the quality difference cause of the Internet of Things node according to the quality difference cause corresponding to the quality difference index of the Internet of Things node.

[0119] The second determination module 630 is configured to retrieve the second subgraph according to the quality difference cause of the IoT node to determine the quality difference coping operation of the IoT node.

[0120] Among them, the second subgraph is the relationship subgraph between the causes of poor quality and the operations to deal with poor quality in the poor quality knowledge graph of the Internet of Things.

[0121] The second determination module 630 searches the relationship subgraph between the quality difference reason and the quality difference coping operation according to the quality difference reason of the IoT node, and obtains the quality difference coping operation corresponding to the quality difference reason of the IoT node. Then, the second determination module 630 determines the quality difference coping operation of the IoT node according to the quality difference coping operation corresponding to the quality difference reason of the IoT node. For example, the quality difference coping operation corresponding to the quality difference reason of the IoT node is used as the quality difference coping operation of the IoT node.

[0122] In the disclosed embodiments, the above device can realize automatic diagnosis and processing of poor quality of the Internet of Things, improve the efficiency and accuracy of poor quality diagnosis and troubleshooting of the Internet of Things, effectively save labor costs, and assist or even replace front-line operation and maintenance personnel to complete the rapid and accurate processing of faults. In addition, compared with poor quality diagnosis and troubleshooting based on neural network models, the poor quality diagnosis and processing of the Internet of Things in the disclosed embodiments is more explainable and conforms to the usage habits of front-line operation and maintenance personnel.

[0123] Figure 7 FIG. 1 is a block diagram showing an IoT quality difference processing device according to other embodiments of the present disclosure. Figure 7 As shown, the Internet of Things quality difference processing device 700 includes a memory 710; and a processor 720 coupled to the memory 710. The memory 710 is used to store instructions for executing the Internet of Things quality difference processing method corresponding to the embodiment. The processor 720 is configured to execute the Internet of Things quality difference processing method in any of some embodiments of the present disclosure based on the instructions stored in the memory 710.

[0124] Figure 8 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

[0125] like Figure 8 As shown, the computer system 800 may be embodied in the form of a general-purpose computing device. The computer system 800 includes a memory 810, a processor 820, and a bus 830 that connects various system components.

[0126] The memory 810 may include, for example, a system memory, a non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application, a boot loader, and other programs. The system memory may include a volatile storage medium, such as a random access memory (RAM) and / or a cache memory. The non-volatile storage medium may store, for example, instructions for executing at least one of the corresponding embodiments of the Internet of Things quality difference processing method. The non-volatile storage medium includes, but is not limited to, a disk memory, an optical memory, a flash memory, and the like.

[0127] The processor 820 can be implemented by a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistors, etc. Discrete hardware components. Accordingly, each module such as the acquisition module, the first determination module, and the second module can be implemented by a central processing unit (CPU) running instructions in a memory that execute corresponding steps, or can be implemented by a dedicated circuit that executes corresponding steps.

[0128] The bus 830 may use any of a variety of bus architectures, including, but not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, and a Peripheral Component Interconnect (PCI) bus.

[0129] The computer system 800 may also include an input / output interface 840, a network interface 850, a storage interface 860, etc. These interfaces 840, 850, 860, the memory 810, and the processor 820 may be connected via a bus 830. The input / output interface 840 may provide a connection interface for input / output devices such as a display, a mouse, and a keyboard. The network interface 850 may provide a connection interface for various networked devices. The storage interface 860 may provide a connection interface for external storage devices such as a floppy disk, a USB flash drive, and an SD card.

[0130] Here, various aspects of the present disclosure are described with reference to flowcharts and / or block diagrams of methods, devices, and computer program products according to embodiments of the present disclosure. It should be understood that each frame of the flowchart and / or block diagram and the combination of frames can be implemented by computer-readable program instructions.

[0131] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable device to produce a machine, so that the processor executes the instructions to produce means for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0132] These computer-readable program instructions may also be stored in a computer-readable memory, which cause the computer to work in a specific manner to produce an article of manufacture, including instructions for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0133] The present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0134] The Internet of Things poor quality processing method and device in the above-mentioned embodiments can improve the efficiency and accuracy of Internet of Things poor quality diagnosis and troubleshooting, and assist or even replace front-line operation and maintenance personnel to complete the rapid and accurate processing of faults.

[0135] So far, the method and device for processing poor quality of the Internet of Things according to the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

Claims

1. A method for processing poor quality of Internet of Things, comprising: Get the quality index of IoT nodes; Retrieving a first subgraph according to the poor quality index of the IoT node to determine the poor quality cause of the IoT node, wherein the first subgraph is a relationship subgraph between the poor quality index and the poor quality cause in the poor quality knowledge graph of the IoT; Retrieving a second subgraph according to the quality difference cause of the IoT node to determine the quality difference coping operation of the IoT node, wherein the second subgraph is a relationship subgraph between the quality difference cause and the quality difference coping operation in the quality difference knowledge graph of the IoT; Wherein, searching the first subgraph according to the quality difference index of the IoT node to determine the quality difference cause of the IoT node includes: In a case where there is no common poor quality cause corresponding to all the poor quality indicators of the Internet of Things nodes in the first subgraph, determining a plurality of candidate poor quality causes according to the poor quality causes corresponding to all the poor quality indicators of the Internet of Things nodes; Determining a recommended score for each of the plurality of candidate poor quality reasons, comprising: determining an associated poor quality indicator for each candidate poor quality reason among all the poor quality indicators of the IoT node; determining a relationship weight between each candidate poor quality reason and its associated poor quality indicator; determining a recommended score for each candidate poor quality reason according to the relationship weight between each candidate poor quality reason and its associated poor quality indicator; The poor quality cause of the Internet of Things node is determined according to the recommendation score of each of the multiple candidate poor quality causes.

2. The method for processing poor quality of the Internet of Things according to claim 1, wherein: Retrieving the first subgraph according to the quality index of the IoT node to determine the quality cause of the IoT node also includes: In a case where there is a common poor quality cause corresponding to all the poor quality indicators of the Internet of Things node in the first subgraph, the poor quality cause of the Internet of Things node is determined according to the common poor quality cause.

3. The method for processing poor quality of the Internet of Things according to claim 1, wherein: Determining the relationship weight between each candidate poor quality reason and its associated poor quality index includes: In the case where the associated poor quality indicator is directly associated with the candidate poor quality cause, the relationship weight between the candidate poor quality cause and the associated poor quality indicator is determined according to the weight of the edge between the candidate poor quality cause and the associated poor quality indicator in the first subgraph.

4. The method for processing poor quality of the Internet of Things according to claim 3, wherein: Determining the relationship weight between each candidate poor quality reason and its associated poor quality index also includes: In the case where the associated poor quality indicator is indirectly associated with the candidate poor quality cause through an intermediate poor quality indicator, the relationship weight between the candidate poor quality cause and the associated poor quality indicator is determined according to the weight of the edge between the candidate poor quality cause and the intermediate poor quality indicator in the first subgraph, and the weight of the edge between the intermediate poor quality indicator and the associated poor quality indicator.

5. The method for processing poor quality of the Internet of Things according to claim 1, wherein: The quality index of obtaining the IoT node includes: Receive IoT poor quality processing request; The quality difference index of the Internet of Things node is obtained from the Internet of Things quality difference processing request.

6. The method for processing poor quality of the Internet of Things according to claim 1, wherein: The quality difference knowledge graph of the Internet of Things also includes: a relationship subgraph between the Internet of Things node type and the quality difference type, a relationship subgraph between the Internet of Things node type and the quality difference index, a relationship subgraph between the quality difference type and the quality difference index, and a relationship subgraph between the quality difference type and the quality difference cause.

7. The method for processing poor quality of the Internet of Things according to claim 1, further comprising: Generate a work order according to the poor quality cause of the IoT node and the poor quality response operation of the IoT node; The work order is sent to a user terminal.

8. The method for processing poor quality of the Internet of Things according to claim 7, further comprising: According to the quality difference processing result data fed back by the user terminal, the quality difference knowledge graph of the Internet of Things is updated.

9. The method for processing poor quality of the Internet of Things according to claim 1, further comprising: Construct an ontology model of the quality difference knowledge graph of the Internet of Things; According to the ontology model of the quality difference knowledge graph of the Internet of Things, a quality difference knowledge graph of the Internet of Things is constructed.

10. An IoT quality-poor processing device, comprising: An acquisition module, configured to acquire a quality index of an IoT node; A first determination module is configured to retrieve a first subgraph according to the quality difference index of the IoT node to determine the quality difference cause of the IoT node, wherein the first subgraph is a relationship subgraph between the quality difference index and the quality difference cause in the quality difference knowledge graph of the IoT; A second determination module is configured to retrieve a second subgraph according to the quality difference cause of the IoT node to determine the quality difference coping operation of the IoT node, wherein the second subgraph is a relationship subgraph between the quality difference cause and the quality difference coping operation in the quality difference knowledge graph of the IoT; The first determining module retrieves the first subgraph according to the quality index of the IoT node to determine the quality cause of the IoT node, including: In a case where there is no common poor quality cause corresponding to all the poor quality indicators of the Internet of Things nodes in the first subgraph, determining a plurality of candidate poor quality causes according to the poor quality causes corresponding to all the poor quality indicators of the Internet of Things nodes; Determining a recommended score for each of the plurality of candidate poor quality reasons, comprising: determining an associated poor quality indicator for each candidate poor quality reason among all the poor quality indicators of the IoT node; determining a relationship weight between each candidate poor quality reason and its associated poor quality indicator; determining a recommended score for each candidate poor quality reason according to the relationship weight between each candidate poor quality reason and its associated poor quality indicator; The poor quality cause of the Internet of Things node is determined according to the recommendation score of each of the multiple candidate poor quality causes.

11. An IoT quality-poor processing device, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the Internet of Things quality difference processing method as described in any one of claims 1 to 9 based on instructions stored in the memory.

12. A computer storable medium having computer program instructions stored thereon, which, when executed by a processor, implements the Internet of Things quality difference processing method as described in any one of claims 1 to 9.