A network operation and maintenance method, device and system

Abstract

The embodiments of the present application provide a network operation and maintenance method, device and system for solving the problems of high cost and low operation and maintenance efficiency of human-based network operation and maintenance methods. The method includes: a network intelligent unit obtains first network data, and performs training based on the first network data to obtain a data model, and deploys the data model. The data model is a first model, a second model or a third model, the first model is used to determine the network control instructions sent to the operation support system, the second model is used to determine the network control instructions sent to the network element / network management system, and the third model is used to determine the network control instructions sent to the network element. The network intelligent unit performs AI training on various uploaded network data to generate a data model, and the generated model is deployed to the inference unit. The inference unit performs on-demand real-time data model inference, which can improve the real-time intelligence level of the network.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a network operation and maintenance method, device and system. Background Art

[0002] With the development of communications, telecommunication networks are becoming more and more complex. For example, wireless 2G / 3G / 4G / 5G forms an overlay network of multiple wireless networks, and core networks such as CS / PS / IMS / IOT form a network structure in which multiple core networks coexist.

[0003] At present, the operation and maintenance management of telecommunications networks is mainly people-oriented. According to the data analysis of more than 1,800 typical operation and maintenance activities, 95% of the processes and operation nodes require human intervention. For example, in the home broadband complaint handling process, 15 process nodes and more than 100 operation nodes rely on manual analysis and decision-making and isolated auxiliary tools, which leads to the need to maintain a large operation and maintenance team.

[0004] The human-centered operation and maintenance management model leads to high costs. Moreover, the human-centered operation and maintenance management model relies on processes and manual transmission between upstream and downstream, and relies on the personal knowledge base of experts, which is inefficient. Summary of the invention

[0005] The embodiments of the present application provide a network operation and maintenance method, device and system for solving the problems of high cost and low operation and maintenance efficiency of human-based network operation and maintenance methods.

[0006] In the first aspect, the present application provides a network operation and maintenance method, the method comprising: a network intelligent unit obtains first network data, and performs training based on the first network data to obtain a data model, wherein the data model is a first model, a second model or a third model, the first model is used to determine the network control instructions sent to the operation support system (OSS), the second model is used to determine the network control instructions sent to the element / network management system (EMS / NMS), and the third model is used to determine the network control instructions sent to the network element. The network intelligent unit deploys the data model. In the network operation and maintenance system provided in the embodiment of the present application, the network intelligent unit can perform AI training on various uploaded network data to generate a data model, and deploy the generated model to the cross-domain control unit, the autonomous domain control unit and the network element control unit, and the cross-domain control unit, the autonomous domain control unit and the network element control unit perform on-demand real-time data model reasoning and knowledge reasoning, thereby improving the real-time intelligence level of the network.

[0007] In one possible design, the network intelligence unit deploys the data model, including: the network intelligence unit determines the inference unit of the data model to be deployed. The network intelligence unit sends the data model to the inference unit. In the above design, the network intelligence unit can implement a cloud training model and send it to the inference unit in the telecommunications network.

[0008] In one possible design, the data model is a first model. The network intelligent unit determines the reasoning unit of the data model to be deployed, including: the network intelligent unit determines that the reasoning unit of the data model to be deployed is the reasoning unit included in the cross-domain control unit, and the cross-domain control unit is used to send network control instructions to the OSS according to the reasoning result, or the cross-domain control unit is deployed in the OSS. The network intelligent unit sends the data model to the reasoning unit, including: the network intelligent unit sends the first model to the reasoning unit included in the cross-domain control unit. The above design can support a cross-domain intelligent operation and maintenance closed loop by deploying a cross-domain control unit including a cross-domain reasoning unit in a telecommunications network.

[0009] In one possible design, the data model is the second model. The network intelligent unit determines the reasoning unit of the data model to be deployed, including: the network intelligent unit determines that the reasoning unit of the data model to be deployed is the reasoning unit included in the autonomous domain control unit, and the autonomous domain control unit is used to send network control instructions to the EMS / NMS according to the reasoning result, or the autonomous domain control unit is deployed in the EMS / NMS; the network intelligent unit sends the data model to the reasoning unit, including: the network intelligent unit sends the second model to the reasoning unit included in the autonomous domain control unit. The above design can support a network-level intelligent operation and maintenance closed loop by deploying an autonomous domain control unit including an autonomous domain reasoning unit in a telecommunications network.

[0010] In one possible design, the data model is a third model; the network intelligent unit determines the inference unit of the data model to be deployed, including: the network intelligent unit determines that the inference unit of the data model to be deployed is the inference unit included in the network element control unit, and the network element control unit is used to send network control instructions to the network element according to the inference result, or the network element control unit is deployed in the network element; the network intelligent unit sends the data model to the inference unit, including: the network intelligent unit sends the third model to the inference unit included in the network element control unit. The above design can support the network element-level intelligent operation and maintenance closed loop by deploying the network element inference unit in the network element.

[0011] In a possible design, after the network intelligence unit deploys the data model, the network intelligence unit receives the second network data reported by the inference unit, the second network data being the network data used to input the data model for inference; the network intelligence unit optimizes the data model based on the second network data; and the network intelligence unit deploys the optimized data model. Through the above design, the network intelligence unit can continuously optimize the data model, thereby improving the accuracy of the data model.

[0012] In the second aspect, the present application provides a network operation and maintenance method, the method comprising: an inference unit receives a data model sent by a network intelligent unit, the data model is a first model or a second model or a third model, wherein the first model is used to determine the network control instruction sent to the OSS, the second model is used to determine the network control instruction sent to the EMS / NMS, and the third model is used to determine the network control instruction sent to the network element; the inference unit obtains the second network data, the second network data is the network data used to obtain the inference result; the inference unit inputs the second network data into the data model for inference to obtain the network control instruction; the inference unit executes the network control instruction. In the present application, the inference unit can download and run the data model trained by the network intelligent unit, and complete real-time inference locally, and return the result of the data model inference to the requesting program to execute the control of the network behavior, so as to achieve the network management such as the processing of network events, the adjustment of network resources, and the adjustment of energy consumption, thereby improving the real-time intelligence level of the network.

[0013] In one possible design, the inference unit inputs the second network data into the data model, including: the inference unit obtains third network data, the third network data is the network data obtained by the inference unit; the inference unit performs local optimization on the data model based on the third network data; the inference unit uses the optimized data model for inference to obtain the network control command. In the above implementation, by further optimizing the data model based on the local network data, the accuracy of the data model can be improved, thereby improving the accuracy of the network control instruction.

[0014] In one possible design, the data model is the first model, the reasoning unit is deployed in the cross-domain control unit, and the cross-domain control unit is used to send network control instructions to the OSS according to the reasoning result, or the cross-domain control unit is deployed in the OSS. The above design can support a cross-domain intelligent operation and maintenance closed loop by deploying a cross-domain control unit including a cross-domain reasoning unit in the telecommunications network.

[0015] In a possible design, the reasoning unit obtains the second network data, including: the reasoning unit obtains the second network data from the OSS; or the reasoning unit obtains the second network data from the autonomous domain control unit connected to the cross-domain control unit; or the reasoning unit obtains the second network data from the EMS / NMS connected to the cross-domain control unit. Through the above design, the reasoning unit can obtain network data sent by multiple EMS / NMSs managed by the same OSS, thereby realizing cross-EMS / NMS network management and cross-EMS / NMS collaboration.

[0016] In one possible design, the reasoning unit executes the network control instruction, including: the reasoning unit calls the interface of the OSS to execute the network control instruction; or the reasoning unit sends the network control instruction to the autonomous domain control unit connected to the cross-domain control unit, and the autonomous domain control unit is used to send the network control instruction to the EMS / NMS. Through the above design, the reasoning unit can implement cross-EMS / NMS network management and cross-EMS / NMS collaboration.

[0017] In one possible design, the data model is the second model, the reasoning unit is deployed in the autonomous domain control unit, and the autonomous domain control unit is used to send network control instructions to the EMS / NMS according to the reasoning result, or the autonomous domain control unit is deployed in the EMS / NMS. The above design can support a network-level intelligent operation and maintenance closed loop by deploying an autonomous domain control unit including an autonomous domain reasoning unit in a telecommunications network.

[0018] In a possible design, the reasoning unit obtains the second network data, including: the reasoning unit obtains the second network data from the EMS / NMS; or, the reasoning unit obtains the second network data from the network element control unit connected to the autonomous domain control unit; or, the reasoning unit obtains the second network data from the network element connected to the autonomous domain control unit. Through the above design, the reasoning unit can obtain network data sent by multiple network elements managed by the same EMS / NMS, thereby realizing cross-network network management and cross-network collaboration.

[0019] In a possible design, the reasoning unit executes the network control instruction, including: the reasoning unit calls the interface of the EMS / NMS to execute the network control instruction; or the reasoning unit sends the network control instruction to at least one network element control unit connected to the autonomous domain control unit, and the network element control unit is used to send the network control instruction to the network element. Through the above design, the reasoning unit can realize cross-network network management and cross-network collaboration.

[0020] In a possible design, the data model is the third model, the reasoning unit is deployed in the network element control unit, and the network element control unit is used to send network control instructions to the network element according to the reasoning result, or the network element control unit is deployed in the network element. The above design can support a network element-level intelligent operation and maintenance closed loop by deploying the network element reasoning unit in the network element.

[0021] In a possible design, the reasoning unit obtains the second network data, including: the reasoning unit obtains the second network data from the network element. Through the above design, the reasoning unit can implement the management of a single network element.

[0022] In a possible design, the reasoning unit executes the network control instruction, including: the reasoning unit calls the interface of the network element to execute the network control instruction. Through the above design, the reasoning unit can implement the management of a single network element.

[0023] In a possible design, after the inference unit inputs the second network data into the data model, the inference unit further includes: sending the second network data to the network intelligence unit, where the second network data includes the second network data and the network control instruction. Through the above design, the network intelligence unit can continuously optimize the data model, thereby improving the accuracy of the data model.

[0024] In a third aspect, the present application provides a network operation and maintenance device, which may be a network intelligent unit, or may be a chip or chipset in a network intelligent unit. The device has the function of implementing any design in the first aspect above. The function may be implemented by hardware, or may be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0025] In a fourth aspect, the present application provides a network operation and maintenance device, which may be an inference unit, or may be a chip or chipset in an inference unit. The device has the function of implementing any design in the second aspect above. The function may be implemented by hardware, or may be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0026] In a fifth aspect, the present application provides a cross-domain control unit, which may include the reasoning unit described in the fourth aspect above, and the reasoning unit is used to store the first model, and input the network data obtained by the cross-domain control unit into the first model for reasoning to obtain network control instructions. The cross-domain control unit can be integrated in the OSS or deployed separately from the OSS. Among them, when the cross-domain control unit is deployed separately from the OSS, it can be connected to the OSS for communication, and can be used to send network control instructions to the OSS based on the reasoning results.

[0027] In a sixth aspect, the present application provides an autonomous domain control unit, which may include the reasoning unit described in the fourth aspect above, and the reasoning unit is used to store the second model, and input the network data obtained by the autonomous domain control unit into the second model for reasoning to obtain network control instructions. The autonomous domain control unit can be integrated into the EMS / NMS, or it can be deployed separately from the EMS / NMS. Among them, when the autonomous domain control unit is deployed separately from the EMS / NMS, it can be connected to the EMS / NMS for communication, and can be used to send network control instructions to the EMS / NMS based on the reasoning results.

[0028] In a seventh aspect, the present application provides a network element control unit, which may include the reasoning unit described in the fourth aspect above, and the reasoning unit is used to store a third model, and input the network data obtained by the network element control unit into the third model for reasoning to obtain network control instructions. The network element control unit can be integrated in the network element, or it can be deployed separately from the network element. Among them, when the network element control unit is deployed separately from the network element, it can be connected to the network element for communication, and can be used to send network control instructions to the network element according to the reasoning result.

[0029] In an eighth aspect, the present application also provides a network operation and maintenance system, which includes the network operation and maintenance device described in the third aspect and the network operation and maintenance device described in the fourth aspect.

[0030] In one possible design, the system may also include OSS.

[0031] In one possible design, the system may also include an EMS / NMS.

[0032] In one possible design, the system may also include network infrastructure equipment, such as network elements.

[0033] In one possible design, the system may also include a business support system (BSS).

[0034] In a ninth aspect, the present application also provides a network operation and maintenance system, which includes the network operation and maintenance device described in the third aspect and the cross-domain control unit described in the fifth aspect.

[0035] In one possible design, the system may also include OSS.

[0036] In one possible design, the system may also include an EMS / NMS.

[0037] In one possible design, the system may also include network infrastructure equipment, such as network elements.

[0038] In one possible design, the system may also include a BSS.

[0039] In one possible design, the system may also include the autonomous domain management and control unit described in the sixth aspect above.

[0040] In one possible design, the system may also include the network element management and control unit described in the seventh aspect above.

[0041] In the tenth aspect, the present application also provides a network operation and maintenance system, which includes the network operation and maintenance device described in the third aspect and the autonomous domain management and control unit described in the sixth aspect.

[0042] In one possible design, the system may also include OSS.

[0043] In one possible design, the system may also include an EMS / NMS.

[0044] In one possible design, the system may also include network infrastructure equipment, such as network elements.

[0045] In one possible design, the system may also include a BSS.

[0046] In a possible design, the system may also include the cross-domain management and control unit described in the fifth aspect above.

[0047] In one possible design, the system may also include the network element management and control unit described in the seventh aspect above.

[0048] In the eleventh aspect, the present application also provides a network operation and maintenance system, which includes the network operation and maintenance device described in the third aspect and the network element management and control unit described in the sixth aspect.

[0049] In one possible design, the system may also include OSS.

[0050] In one possible design, the system may also include an EMS / NMS.

[0051] In one possible design, the system may also include network infrastructure equipment, such as network elements.

[0052] In one possible design, the system may also include a BSS.

[0053] In one possible design, the system may also include the autonomous domain management and control unit described in the sixth aspect above.

[0054] In a possible design, the system may also include the cross-domain management and control unit described in the fifth aspect above.

[0055] In a twelfth aspect, the present application further provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the methods described in the above aspects.

[0056] In a thirteenth aspect, the present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the methods described in the above aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of the architecture of a telecommunications network provided in an embodiment of the present application;

[0058] Figure 2 A schematic diagram of an operation and maintenance process provided in an embodiment of the present application;

[0059] Figure 3 A schematic diagram of the architecture of a network management system provided in an embodiment of the present application;

[0060] Figure 4 A schematic diagram of the structure of another wireless access network device provided in an embodiment of the present application;

[0061] Figure 5 A schematic diagram of a process for building a model provided in an embodiment of the present application;

[0062] Figure 6 A schematic diagram of an AI reasoning process provided in an embodiment of the present application;

[0063] Figure 7 A schematic diagram of an automated operation and maintenance process provided in an embodiment of the present application;

[0064] Figure 8 A schematic diagram of an automated operation and maintenance process provided in an embodiment of the present application;

[0065] Fig. 9 A schematic diagram of an automated operation and maintenance process provided in an embodiment of the present application;

[0066] Fig.10 A schematic diagram of an automated operation and maintenance process provided in an embodiment of the present application;

[0067] Fig.11 A schematic diagram of an automated operation and maintenance process provided in an embodiment of the present application;

[0068] Fig.12 A schematic diagram of the structure of a network operation and maintenance device provided in an embodiment of the present application;

[0069] Fig.13 A schematic diagram of the structure of another network operation and maintenance device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0071] With the development of communications, telecommunication networks are becoming more and more complex. For example, wireless 2G / 3G / 4G / 5G forms an overlay network where multiple networks coexist, and core networks such as CS / PS / IMS / IOT form a network pattern where multiple network domains coexist. At present, the operation and maintenance management of telecommunication networks is mainly people-oriented. According to the data analysis of more than 1,800 typical operation and maintenance activities, 95% of the processes and operation nodes require human intervention, and how to improve the operation and maintenance efficiency has become the focus. For example, in the process of handling complaints about home broadband, 15 process nodes and more than 100 operation nodes rely on manual analysis and decision-making and isolated auxiliary tools, resulting in the need to maintain a large operation and maintenance team.

[0072] Currently, the network operation and maintenance system can include BSS, OSS, EMS / NMS and network elements / networks, e.g. Figure 1 As shown in the figure, BSS is a system for customer relationship management, business supply chain management, and business decision support. OSS is a system for running and monitoring the network. EMS / NMS is used to manage network infrastructure. Network elements / networks are network infrastructure, such as routers, base stations, switches, etc. At present, network operation, maintenance, and management are mainly completed by OSS / NMS / EMS in collaboration.

[0073] The network element / network is the network infrastructure, and the issuance of instructions to it is mainly completed by NMS / EMS. NMS / EMS is responsible for monitoring and managing the network element / network, and reporting operation and maintenance information to OSS. OSS manages the network based on the operation and maintenance information. At present, OSS mainly relies on people to manage the network, such as manual monitoring of network status, manual dispatching, etc. Before dispatching, there are still a large number of auxiliary operation and maintenance processes, such as delimitation, positioning, plan formulation, review, verification, and audit, which need to be executed by network management personnel. After receiving the report of operation and maintenance information, OSS executes dispatching through a series of operation and maintenance processes, and the network management personnel execute the instructions to the network / network element on NMS / EMS. For example, the network management process can be as follows Figure 2 shown.

[0074] In the telecommunications network management and operation and maintenance system, a large number of processes and operation nodes require human intervention. The human-dominated operation and maintenance mode leads to low operation and maintenance efficiency. The lack of intelligent analysis of massive data makes it difficult to accurately locate and quickly recover network events, and the utilization rate of network resources is low and energy consumption is high. In addition, the current human-dominated operation and maintenance management model leads to high costs.

[0075] Based on this, the embodiments of the present application provide a network operation and maintenance method, device and system, which are used to solve the problems of low operation and maintenance efficiency, low utilization of network resources and high energy consumption caused by the human-based operation and maintenance method in the prior art. Among them, the method and the device are based on the same inventive concept. Since the principles of solving the problems by the method and the device are similar, the implementation of the device and the method can refer to each other, and the repeated parts will not be repeated.

[0076] The term "plurality" in this application refers to two or more than two. The term "at least one" in this application refers to one or more than one, including one, two, three and more.

[0077] In addition, it should be understood that, in the description of this application, words such as "first" and "second" are only used for the purpose of distinguishing the description, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying an order.

[0078] An embodiment of the present application provides a network management system, which includes a network intelligence unit, wherein the network intelligence unit can be used to perform data management and train data models for network management, etc. In one implementation, the data model can be trained using artificial intelligence (AI).

[0079] In one implementation, the network intelligence unit may include a design module, a training module, and a data lake, wherein the design module may be used to design a data model, the training module may be used to train a data model, and the data lake may store network data reported by each OSS, wherein the network data reported by the OSS may include network data generated by the EMS / NMS managed by the OSS, the network data generated by the EMS / NMS may include network data generated by the EMS / NMS itself, and may also include network data generated by network elements managed by the EMS / NMS. Exemplarily, a data lake may refer to a system that stores data in a natural format such as a large binary object or file. The data lake may store network data in a unified manner, and the data lake may include both raw data generated by the network and converted data, such as data converted for reporting, visualization, data analysis, and machine learning. The data lake may include, but is not limited to, structured data, semi-structured data, unstructured data, and binary data of a relational database.

[0080] The system may also include at least one of the following units: a cross-domain management and control unit, an autonomous domain management and control unit, and a network element management and control unit.

[0081] Among them, the cross-domain control unit can be connected to the OSS for communication, and can be used to send network control instructions to the OSS according to the reasoning results. The cross-domain control unit can realize cross-EMS / NMS network management and cross-EMS / NMS collaboration. In an exemplary description, the cross-domain control unit may include a reasoning unit, which may determine the network control instructions based on the network data sent by the OSS or the network data sent by the EMS / NMS managed by the OSS. In some embodiments, the cross-domain control unit may be integrated in the OSS or deployed separately from the OSS.

[0082] In one implementation, the cross-domain management and control unit may also include an AI application and a planning, construction, maintenance and optimization module, wherein the AI ​​application can be used to solve one or more network problems (for example, dual connectivity (DC) energy-saving applications, high-definition voice call (voice over LTE, VoLTE) voice risk prevention, etc.), and the planning, construction, maintenance and optimization module can be used to plan, build, maintain and optimize the network.

[0083] The autonomous domain control unit can be connected to the EMS / NMS for communication and is used to send control instructions to the EMS / NMS based on the inference results. The autonomous domain control unit can realize network management and collaboration across multiple network elements. In an exemplary description, the autonomous domain control unit may include an inference unit, which can determine network control instructions based on network data sent by the EMS / NMS or network data sent by network elements managed by the EMS / NMS. In some embodiments, the autonomous domain control unit can be integrated into the EMS / NMS or deployed separately from the EMS / NMS.

[0084] In one implementation, the autonomous domain management and control unit may also include an AI application and a network management control analysis module, wherein the AI ​​application can be used to solve one or more network problems, and the network management control analysis module can be used to manage and control the network, as well as to analyze information such as performance alarms generated by the network.

[0085] In some embodiments, the autonomous domain management and control unit may be integrated into the cross-domain management and control unit.

[0086] The network element control unit can be connected to the network element in communication and used to send control instructions to the network element according to the inference result. The network element control unit can implement network management of a single network element. In an exemplary description, the network element control unit can include an inference unit, which can determine at least one network control instruction based on the network data sent by the network element. In some embodiments, the network element control unit can be integrated in the network element or deployed separately from the network element.

[0087] In one implementation, the network element management and control unit may further include a network element sensor, wherein the network element sensor may be used to obtain operation status information of the network element.

[0088] For example, the network operation and maintenance system can be as follows: Figure 3 or Figure 4 As shown, it should be understood that Figure 3 or Figure 4 This is only an exemplary description and does not specifically limit the units included in the network operation and maintenance system, the modules included in each unit, and the deployment location of each unit.

[0089] The present invention provides a network operation and maintenance method, which can be applied to Figure 3 or Figure 4 The network management system shown in the figure may include a model building process of a network intelligence unit and an automated operation and maintenance process of a reasoning unit.

[0090] like Figure 5 As shown, the model building process of the network intelligence unit can be:

[0091] S501, the network intelligence unit obtains first network data, wherein the first network data is used for training to obtain a data model.

[0092] In one implementation, the network intelligence unit may subscribe to and obtain the first network data from the data lake.

[0093] In some embodiments, after acquiring the first network data, the network intelligence unit may pre-process the first network data.

[0094] Exemplarily, the first network data may be network data related to the data model, or may be network data related to the problem solved by the data model. For example, the data model is a model for determining the relevance of network alarms, and the first network data may also be alarm data generated in the network, and so on.

[0095] In some embodiments, before step S501, the network intelligence unit determines the problem to be solved by the data model and data related to the problem. Exemplarily, the data model can be a model for solving the energy saving problem of the data center, or the data model can also be a model for determining the relevance of network alarms.

[0096] In one implementation, the network administrator may define the problem to be solved by the data model and the data related to the problem through the design unit (eg, the network administrator defines the problem through the interface or configuration file provided by the design unit).

[0097] S502: The network intelligence unit performs training based on the first network data to obtain a data model.

[0098] The data model can be a first model, a second model, or a third model, wherein the first model is used to determine the network control instructions sent to the OSS, the second model is used to determine the network control instructions sent to the EMS / NMS, and the third model is used to determine the network control instructions sent to the network element.

[0099] In an exemplary description, if the data model is the first model, the first network data may be network data sent by the OSS or sent by the EMS / NMS managed by the OSS.

[0100] If the data model is the second model, the first network data may be network data sent by the EMS / NMS or network data sent by a network element managed by the EMS / NMS.

[0101] If the data model is the third model, the first network data may be network data sent by the network element.

[0102] In one implementation, a training module of the network intelligence unit may be trained using the first network data.

[0103] S503, the network intelligence unit deploys a data model.

[0104] In one implementation, the network intelligence unit determines the reasoning unit where the data model is to be deployed, and sends the data model to the reasoning unit.

[0105] In an exemplary description, the reasoning unit of the data model to be deployed can be determined by the network management personnel through the design unit before the network intelligence unit trains the data model.

[0106] For example, assuming that the data model is the first model, the inference unit to be deployed for the data model may be the inference unit included in the cross-domain management and control unit. Therefore, when the network intelligence unit deploys the first model, the first model may be sent to the inference unit included in the cross-domain management and control unit.

[0107] For another example, assuming that the data model is the second model, the reasoning unit to be deployed for the data model may be the reasoning unit included in the autonomous domain control unit. Therefore, when the network intelligence unit deploys the second model, the second model may be sent to the reasoning unit included in the autonomous domain control unit.

[0108] For another example, assuming that the data model is the third model, the inference unit to be deployed for the data model may be the inference unit included in the network element management and control unit. Therefore, when the network intelligence unit deploys the third model, the third model may be sent to the inference unit included in the network element management and control unit.

[0109] like Figure 6 As shown in the figure, the automated operation and maintenance process of the inference unit can be:

[0110] S601, the inference unit receives a data model sent by the network intelligence unit, where the data model is the first model, the second model, or the third model.

[0111] In an exemplary description, if the reasoning unit is deployed in a cross-domain control unit, the data model may be a first model. If the reasoning unit is deployed in an autonomous domain control unit, the data model may be a second model. If the reasoning unit is deployed in a network element control unit, the data model may be a third model.

[0112] S602: The inference unit obtains second network data, where the second network data is network data used to obtain an inference result.

[0113] In a possible implementation, if the inference unit is deployed in the cross-domain control unit, the inference unit can obtain the second network data in any of the following three ways:

[0114] Method 1: The inference unit obtains the second network data from the OSS, where the second network data is network data sent by the OSS or sent by the EMS / NMS managed by the OSS.

[0115] Method 2: The inference unit obtains the second network data from the autonomous domain control unit connected to the cross-domain control unit, where the second network data is the network data sent by the EMS / NMS connected to the autonomous domain control unit.

[0116] Method three: the inference unit obtains the second network data from the EMS / NMS to which the cross-domain management and control unit is connected, and the second network data includes the network data sent by the EMS / NMS.

[0117] In another possible implementation, if the inference unit is deployed in the autonomous domain control unit, the inference unit can obtain the second network data in any of the following three ways:

[0118] Method 1: The inference unit obtains the second network data from the EMS / NMS, where the second network data includes network data sent by the EMS / NMS or network data sent by a network element managed by the EMS / NMS.

[0119] Method 2: The inference unit obtains the second network data from the network element control unit connected to the autonomous domain control unit, where the second network data includes network data sent by the network element connected to the network element control unit;

[0120] Method three: the inference unit obtains the second network data from the network element to which the autonomous domain control unit is connected.

[0121] In another possible implementation, if the inference unit is deployed in the network element management and control unit, the inference unit may obtain the second network data in the following manner: obtain the second network data from the network element, where the second network data includes the network data sent by the network element.

[0122] In some embodiments, the reasoning unit may send the second network data to the network intelligence unit, so that the data lake of the network intelligence unit may be enriched.

[0123] S603, the inference unit inputs the second network data into the data model for inference to obtain a network control instruction, wherein the network control instruction is determined by the inference result of the data model, or the network control instruction is the inference result of the data model.

[0124] In one implementation, the reasoning unit may first obtain the third network data, and locally optimize the data model based on the third network data. The reasoning unit uses the optimized data model to perform reasoning to obtain the network control command. The third network data may be the network data obtained by the reasoning unit.

[0125] Exemplarily, the third network data may be local network data. For example, assuming that the reasoning unit is deployed in a cross-domain control unit, the third network data may be network data of an OSS connected to the cross-domain control unit. For example, assuming that the reasoning unit is deployed in an autonomous domain control unit, the third network data may be network data of an EMS / NMS connected to the autonomous domain control unit.

[0126] In the above implementation, by further optimizing the data model based on local network data, the accuracy of the data model can be improved, thereby improving the accuracy of the network control instructions.

[0127] S604, the reasoning unit executes the network control instruction, such as the recommended operation for optimal energy saving, or determines the possible fault according to the current alarm.

[0128] In a possible implementation, if the inference unit is deployed in the cross-domain control unit, the inference unit can execute the network control instruction in any of the following two ways:

[0129] Method 1: The inference unit calls the OSS interface to execute network control instructions;

[0130] Method 2: The inference unit sends a network control instruction to the autonomous domain control unit connected to the cross-domain control unit.

[0131] In another possible implementation, the data model is the second model, and the reasoning unit is deployed in the autonomous domain control unit. The reasoning unit can execute the network control instruction in any of the following two ways:

[0132] Method 1: The inference unit calls the EMS / NMS interface to execute network control instructions.

[0133] Method 2: The reasoning unit sends a network control instruction to the network element control unit connected to the autonomous domain control unit.

[0134] In another possible implementation, the data model is the third model, and the inference unit is deployed in the network element management and control unit. The inference unit can execute the network control instruction in the following manner: calling the interface of the network element to execute the network control instruction.

[0135] In a possible implementation, after executing the AI ​​reasoning process, the reasoning unit may send second network data to the network intelligence unit, where the second network data includes the second network data and the network control instructions determined by the data model according to the second network data. The network intelligence unit optimizes the data model based on the second network data. In the above manner, the network intelligence unit can continuously update the data model, thereby improving the accuracy of the network control instructions.

[0136] In order to better understand the network operation and maintenance method provided in the embodiments of the present application, the network management process is described in detail below in combination with the deployment scenarios. It should be understood that the deployment scenarios listed below are only exemplary descriptions and do not specifically limit the network operation and maintenance method provided in the present application.

[0137] In order to distinguish the inference units in the cross-domain control unit, the autonomous domain control unit, and the network element control unit, the inference unit in the cross-domain control unit is referred to as the cross-domain inference unit, the inference unit in the autonomous domain control unit is referred to as the autonomous domain inference unit, and the inference unit in the network element control unit is referred to as the network element inference unit. In order to distinguish the AI ​​applications in the cross-domain control unit, the autonomous domain control unit, and the network element control unit, the AI ​​applications in the cross-domain control unit are referred to as the cross-domain AI applications, the AI ​​applications in the autonomous domain control unit are referred to as the autonomous domain AI applications, and the AI ​​applications in the network element control unit are referred to as the network element AI applications.

[0138] Deployment scenario 1: Deploy cross-domain control units, autonomous domain control units, and network element control units in the telecommunications network. The network element control unit can be integrated in the network element, such as Figure 7 shown.

[0139] Based on the network operation and maintenance system shown in deployment scenario 1, three automated operation and maintenance processes can be run, namely, the automated operation and maintenance process applied to a single network element, the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS, and the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS. The following describes the three automated operation and maintenance processes respectively.

[0140] 1. The automated operation and maintenance process applied to a single network element may include:

[0141] Obtaining the model process:

[0142] 1.1) The network element reasoning unit collects the network data required by the third model locally.

[0143] Among them, the network element reasoning unit may include first network data for training the third model, may also include second network data for inputting the third model to obtain reasoning results, and may also include third network data for optimizing the third model by the network element reasoning unit.

[0144] 1.2) The network element reasoning unit reports the collected network data to the network intelligence unit, and receives the trained third model sent by the network intelligence unit.

[0145] AI reasoning process:

[0146] 1.3) The network element reasoning unit optimizes the third model based on the third network data, and inputs the second network data into the optimized third model for reasoning to obtain a network control instruction.

[0147] 1.4) The network element AI application obtains the network control instruction from the network element reasoning unit and automatically calls the network element function to implement the network control instruction.

[0148] The above-mentioned automated operation and maintenance process applied to a single network element can realize intelligent operation and maintenance of a single network element, thereby improving the operation and maintenance speed and efficiency of a single network element. Especially when the operated service is very sensitive to latency, the above-mentioned automated operation and maintenance process applied to a single network element can reduce the latency caused by operation and maintenance. For example, applications can be intelligently identified and accelerated in home broadband access network elements, thereby improving service experience.

[0149] 2. The automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS may include:

[0150] Obtaining the model process:

[0151] 2.1) The autonomous domain reasoning unit obtains the network data required by the second model.

[0152] For example, 2.1a) the autonomous domain reasoning unit may obtain the network data required by the second model from the interface of the EMS / NMS.

[0153] Alternatively, 2.1b) the autonomous domain reasoning unit may also obtain the network data required by the second model from the network element.

[0154] Among them, the autonomous domain reasoning unit may include first network data for training the second model, may also include second network data for inputting the second model to obtain reasoning results, and may also include third network data for optimizing the second model by the autonomous domain reasoning unit.

[0155] 2.2) The autonomous domain reasoning unit reports the acquired network data to the network intelligence unit, and receives the trained second model sent by the network intelligence unit.

[0156] AI reasoning process:

[0157] 2.3) The autonomous domain reasoning unit optimizes the second model based on the third network data, and inputs the second network data into the optimized second model for reasoning to obtain a network control instruction.

[0158] 2.4) The autonomous domain AI application obtains network control instructions from the autonomous domain reasoning unit and implements the network control instructions.

[0159] For example, 2.4a) autonomous domain AI applications can call the open interface of EMS / NMS to implement network control instructions.

[0160] Alternatively, 2.4b) the autonomous domain AI application may also send network control instructions to the network element management and control unit so that the network element implements the network control instructions.

[0161] Through the above-mentioned automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS, intelligent operation and maintenance of multiple network elements managed by the same EMS / NMS can be realized, thereby improving the operation and maintenance speed and efficiency of multiple network elements. For example, in a wireless network domain, the power and frequency bands of multiple wireless base stations can be adjusted through the above process to achieve better energy efficiency.

[0162] 3. The automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS may include:

[0163] Obtaining the model process:

[0164] 3.1) The cross-domain reasoning unit obtains the network data required by the first model.

[0165] For example, 3.1a) the cross-domain reasoning unit may obtain the network data required by the first model from the interface of the OSS.

[0166] Alternatively, 3.1b) the cross-domain reasoning unit may also obtain the network data required for the first model from the autonomous domain control unit according to the instructions of the cross-domain AI application.

[0167] Alternatively, 3.1c) the cross-domain reasoning unit may also obtain the network data required by the first model from the interface of the EMS / NMS connected to the OSS.

[0168] Among them, the cross-domain reasoning unit may include first network data for training the first model, may also include second network data for inputting the first model to obtain reasoning results, and may also include third network data for optimizing the first model by the cross-domain reasoning unit.

[0169] 3.2) The cross-domain reasoning unit reports the acquired network data to the network intelligence unit, and receives the trained first model sent by the network intelligence unit.

[0170] Reasoning stage:

[0171] 3.3) The cross-domain reasoning unit optimizes the first model based on the third network data, and inputs the second network data into the optimized first model for reasoning to obtain a network control instruction.

[0172] 3.4) The cross-domain AI application obtains network control instructions from the cross-domain reasoning unit and implements the network control instructions.

[0173] For example, 3.4a) cross-domain AI applications can call OSS open interfaces to implement network control instructions.

[0174] Alternatively, 3.4b) the cross-domain AI application may also send network control instructions to the connected autonomous domain control unit so that the autonomous domain control unit implements the network control instructions.

[0175] Through the above-mentioned automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS, intelligent operation and maintenance of multiple EMS / NMS managed by the same OSS can be realized, thereby improving the operation and maintenance speed and efficiency of multiple EMS / NMS. For example, the cause of a wireless base station failure may be an abnormality in the related transmission line or equipment. Through the above-mentioned flow, it is possible to obtain alarm information from the wireless network management and the transmission network management respectively to intelligently associate the alarm correlation.

[0176] In the above deployment scenario one, by deploying network intelligence units in the telecommunications network, it is possible to support training and generating data models in the cloud. By deploying network element reasoning units in network elements, it is possible to support a network element-level intelligent operation and maintenance closed loop. By deploying an autonomous domain control unit including an autonomous domain reasoning unit in the telecommunications network, it is possible to support a network-level intelligent operation and maintenance closed loop. By deploying a cross-domain control unit including a cross-domain reasoning unit in the telecommunications network, it is possible to support a cross-domain intelligent operation and maintenance closed loop. In addition, through the above deployment scenario one, an intelligent operation and maintenance system can be introduced without changing EMS / NMS / OSS / BSS, and intelligent automatic operation and maintenance can be implemented at multiple levels of network elements, networks, and cross-domains, thereby greatly reducing the operation and maintenance manpower cost.

[0177] Deployment scenario 2: Deploy cross-domain control units, autonomous domain control units, and network element control units in the telecommunications network. The cross-domain control unit can be integrated into the OSS, the autonomous domain control unit can be integrated into the EMS / NMS, and the network element control unit can be inherited in the network element, such as Figure 8 shown.

[0178] Based on the network operation and maintenance system shown in deployment scenario 2, three automated operation and maintenance processes can be run, namely, the automated operation and maintenance process applied to a single network element, the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS, and the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS. The following describes the three automated operation and maintenance processes respectively.

[0179] 1. The automated operation and maintenance process applied to a single network element is the same as the process of obtaining the model of the automated operation and maintenance process applied to a single network element in deployment scenario 1, and will not be repeated here.

[0180] 2. The automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS may include:

[0181] Obtaining the model process:

[0182] 2.1) The autonomous domain reasoning unit obtains the network data required by the second model.

[0183] Among them, the autonomous domain reasoning unit may include first network data for training the second model, may also include second network data for inputting the second model to obtain reasoning results, and may also include third network data for optimizing the second model by the autonomous domain reasoning unit.

[0184] In one implementation, the EMS / NMS may obtain the network data required for the second model from the network element. The autonomous domain reasoning unit may obtain the network data required for the second model from the EMS / NMS locally.

[0185] 2.2) The autonomous domain reasoning unit reports the acquired network data to the network intelligence unit, and receives the trained second model sent by the network intelligence unit.

[0186] AI reasoning process:

[0187] 2.3) The autonomous domain reasoning unit optimizes the second model based on the third network data, and inputs the second network data into the optimized second model for reasoning to obtain a network control instruction.

[0188] 2.4) The autonomous domain AI application in EMS / NMS obtains network control instructions from the autonomous domain reasoning unit and implements the network control instructions.

[0189] 3. The automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS may include:

[0190] 3.1) The cross-domain reasoning unit obtains the network data required by the first model.

[0191] In one implementation, the OSS may obtain the network data required for the first model from the EMS / NMS. The cross-domain reasoning unit may obtain the network data required for the first model from the OSS.

[0192] Among them, the cross-domain reasoning unit may include first network data for training the first model, may also include second network data for inputting the first model to obtain reasoning results, and may also include third network data for optimizing the first model by the cross-domain reasoning unit.

[0193] 3.2) The cross-domain reasoning unit reports the acquired network data to the network intelligence unit, and receives the trained first model sent by the network intelligence unit.

[0194] Reasoning stage:

[0195] 3.3) The cross-domain reasoning unit optimizes the first model based on the third network data, and inputs the second network data into the optimized first model for reasoning to obtain a network control instruction.

[0196] 3.4) The cross-domain AI application in OSS obtains network control instructions from the cross-domain reasoning unit and implements the network control instructions.

[0197] The above deployment scenario 2, by deploying network intelligence units in the telecommunications network, can support cloud-based training to generate data models. By adding network element reasoning units to network elements, it is possible to support a network element-level intelligent operation and maintenance closed loop. By adding autonomous domain reasoning units to EMS and NMS, it is possible to support a network-level intelligent operation and maintenance closed loop. By adding cross-domain reasoning units to OSS, it is possible to support a cross-domain intelligent operation and maintenance closed loop. Through the above deployment scenario 2, an intelligent operation and maintenance system can be introduced, so that intelligent and automatic operation and maintenance can be implemented at multiple levels, including network elements, networks, and cross-domains, which can greatly reduce the labor cost of operation and maintenance. In addition, the above deployment scenario 2 can avoid additional management component operation and maintenance costs by upgrading the existing system.

[0198] Deployment scenario 3: Deploy cross-domain control units in the telecommunications network, such as Fig. 9 The cross-domain control unit may be integrated into the OSS or may not be integrated into the OSS.

[0199] Based on the network operation and maintenance system shown in deployment scenario 3, an automated operation and maintenance process can be run, that is, an automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS.

[0200] If the cross-domain control unit is not integrated in OSS, the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS can be found in the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS in deployment scenario 1. It will not be described here.

[0201] If the cross-domain control unit is integrated in OSS, the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS can be found in the automated operation and maintenance process applied to multiple EMS / NMS managed by the same OSS in deployment scenario 2. It will not be described here.

[0202] Deployment scenario 4: Deploy autonomous domain control units and network element control units in the telecommunications network, such as Fig.10 The autonomous domain control unit may be integrated into the EMS / NMS or may not be integrated into the EMS / NMS, and the network element control unit may be integrated into the network element.

[0203] Based on the network operation and maintenance system shown in deployment scenario 4, an automated operation and maintenance process can be run, that is, an automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS.

[0204] If the autonomous domain management and control unit is not integrated in the EMS / NMS, the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS can be specifically referred to in the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS in deployment scenario 1, which will not be repeated here.

[0205] If the autonomous domain control unit is integrated in the EMS / NMS, the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS can be specifically referred to the automated operation and maintenance process applied to multiple network elements managed by the same EMS / NMS in deployment scenario 2, which will not be repeated here.

[0206] Deployment scenario 5: Deployment of network element control units in telecommunications networks, such as Fig.11 The network element management and control unit may be integrated in the network element.

[0207] Based on the network operation and maintenance system shown in deployment scenario 5, an automated operation and maintenance process can be run, that is, an automated operation and maintenance process applied to a single network element.

[0208] For details about the automated operation and maintenance process applied to a single network element, please refer to the automated operation and maintenance process applied to a single network element in deployment scenario 1, which will not be described here.

[0209] In the above five deployment scenarios, the first model, the second model, and the third model can be based on the network intelligence unit. Figure 5 The process shown is obtained by training. For the specific process, please refer to Figure 5 As shown, the repeated parts will not be repeated.

[0210] In the network operation and maintenance system provided in the embodiment of the present application, the network intelligent unit can perform AI training on various uploaded network data to generate data models, and deploy the generated models to the cross-domain control unit, the autonomous domain control unit, and the network element control unit. The cross-domain control unit, the autonomous domain control unit, and the network element control unit perform on-demand real-time data model reasoning and knowledge reasoning, thereby improving the real-time intelligence level of the network. The cross-domain control unit, the autonomous domain control unit, and the network element control unit can download, run, and locally optimize the data model trained by the network intelligent unit, and perform real-time reasoning locally. The cross-domain control unit, the autonomous domain control unit, and the network element control unit can return the results of the data model reasoning to the requesting program to execute the control of the network behavior, so as to achieve network management such as processing of network events, adjustment of network resources, and adjustment of energy consumption.

[0211] In addition, through the collaboration between the network intelligence unit, the cross-domain management and control unit, the autonomous domain management and control unit and the network element management and control unit, the data model and network knowledge can be updated or upgraded, thereby improving the system's adaptability to network changes and upgrades.

[0212] In addition, in the existing network management system, the embodiments of the present application enable machines to intelligently analyze and process network data at different levels, which can improve operation and maintenance efficiency on the one hand, and timely adjust idle network resources and energy consumption on the other hand, thereby improving network resource utilization and energy efficiency.

[0213] Based on the same technical concept as the method embodiment, the embodiment of the present application provides a network operation and maintenance device. The structure of the network operation and maintenance device can be as follows Fig.12 As shown, it includes a communication module 1201 and a processing module 1202.

[0214] In one implementation, the network operation and maintenance device can be used to implement Figures 5 to 11 In the embodiment of the method executed by the network intelligent unit, the device can be the network intelligent unit itself, or it can be a chip or chipset in the network intelligent unit or a part of the chip used to execute the function of the related method. Among them, the communication module 1201 is used to obtain the first network data; the processing module 1202 is used to train based on the first network data obtained by the communication module 1201 to obtain a data model; the data model is a first model, a second model or a third model, the first model is used to determine the network control instructions sent to the OSS, the second model is used to determine the network control instructions sent to the EMS / NMS, and the third model is used to determine the network control instructions sent to the network element; and, deploy the data model.

[0215] The processing module 1202 , when deploying a data model, can be specifically used to: determine an inference unit of the data model to be deployed; and send the data model to the inference unit through the communication module 1201 .

[0216] If the data model is the first model, the processing module 1202, when determining the inference unit of the data model to be deployed, can be specifically used to: determine that the inference unit of the data model to be deployed is an inference unit included in the cross-domain control unit, and the cross-domain control unit is used to send network control instructions to the OSS according to the inference result, or the cross-domain control unit is deployed on the OSS; the processing module 1202, when sending the data model to the inference unit through the communication module 1201, includes: sending the first model to the inference unit included in the cross-domain control unit through the communication module 1201.

[0217] If the data model is the second model, the processing module 1202, when determining the inference unit of the data model to be deployed, is specifically used to: determine that the inference unit of the data model to be deployed is an inference unit included in the autonomous domain control unit, and the autonomous domain control unit is used to send network control instructions to the EMS / NMS according to the inference result, or the autonomous domain control unit is deployed in the EMS / NMS; the processing module 1202, when sending the data model to the inference unit through the communication module 1201, is specifically used to: send the second model to the inference unit included in the autonomous domain control unit through the communication module 1201.

[0218] If the data model is the third model, the processing module 1202, when determining the inference unit of the data model to be deployed, is specifically used to: determine that the inference unit of the data model to be deployed is an inference unit included in the network element management and control unit, and the network element management and control unit is used to send network control instructions to the network element according to the inference result, or the network element management and control unit is deployed in the network element; the processing module 1202, when sending the data model to the inference unit through the communication module 1201, is specifically used to: send the third model to the inference unit included in the network element management and control unit through the communication module 1201.

[0219] The communication module 1201 can also be used to: after the processing module 1202 deploys the data model, receive the second network data reported by the inference unit, the second network data is the network data used to input the data model for inference; the processing module 1202 can also be used to: optimize the data model based on the second network data; and deploy the optimized data model.

[0220] In another implementation, the network operation and maintenance device can be used to implement Figures 5 to 11In the embodiment of the method executed by the reasoning unit, the device can be the reasoning unit itself, or it can be a chip or chipset in the reasoning unit or a part of the chip used to execute the function of the related method. Among them, the communication module 1201 is used to receive the data model sent by the network intelligent unit, and the data model is the first model or the second model or the third model, wherein the first model is used to determine the network control instructions sent to the OSS, the second model is used to determine the network control instructions sent to the network management system EMS / NMS, and the third model is used to determine the network control instructions sent to the network element; the processing module 1202 is used to obtain the second network data, the second network data is the network data used to input the data model for reasoning; and, input the second network data into the data model for reasoning to obtain the network control instructions; and, execute the network control instructions.

[0221] The processing module 1202, when inputting the second network data into the data model, can be specifically used to: obtain third network data, where the third network data is the network data obtained by the reasoning unit; locally optimize the data model based on the third network data; and use the optimized data model for reasoning to obtain a network control command.

[0222] Exemplarily, the data model is a first model, the reasoning unit is deployed in a cross-domain control unit, and the cross-domain control unit is used to send a network control instruction to the OSS according to the reasoning result, or the cross-domain control unit is deployed in the OSS.

[0223] When acquiring the second network data, the processing module 1202 can be specifically used to: acquire the second network data from the OSS; or acquire the second network data from the autonomous domain control unit connected to the cross-domain control unit; or acquire the second network data from the EMS / NMS connected to the cross-domain control unit.

[0224] The processing module 1202, when executing the network control instruction, can be specifically used to: call the interface of the OSS to execute the network control instruction; or, send the network control instruction to the autonomous domain control unit connected to the cross-domain control unit through the communication module 1201, and the autonomous domain control unit is used to send the network control instruction to the EMS / NMS; or, call the interface of the EMS / NMS connected to the cross-domain control unit to execute the network control instruction.

[0225] Exemplarily, the data model is the second model, the reasoning unit is deployed in the autonomous domain control unit, the autonomous domain control unit is used to send network control instructions to the EMS / NMS according to the reasoning result, or the autonomous domain control unit is deployed in the EMS / NMS.

[0226] Processing module 1202, when acquiring the second network data, can be specifically used to: acquire the second network data from EMS / NMS, the second network data including network data sent by EMS / NMS or network data sent by network elements managed by EMS / NMS; or, acquire the second network data from a network element control unit connected to an autonomous domain control unit; or, acquire the second network data from a network element connected to an autonomous domain control unit.

[0227] The processing module 1202, when executing a network control instruction, can be specifically used to: call the interface of the EMS / NMS to execute the network control instruction; or, send the network control instruction to at least one network element control unit connected to the autonomous domain control unit through the communication module 1201, and the network element control unit is used to send the network control instruction to the network element; or, call the interface of the network element connected to the autonomous domain control unit to execute the network control instruction.

[0228] Exemplarily, the data model is the third model, the reasoning unit is deployed in the network element management and control unit, the network element management and control unit is used to send network control instructions to the network element according to the reasoning result, or the network element management and control unit is deployed in the network element.

[0229] The processing module 1202, when acquiring the second network data, may be specifically used to: acquire the second network data from a network element.

[0230] The processing module 1202, when executing the network control instruction, can be specifically used to: call the interface of the network element to execute the network control instruction.

[0231] The communication module may also be used to: after the processing module 1202 inputs the second network data into the data model, send the second network data to the network intelligent unit, where the second network data includes the second network data and the network control instruction.

[0232] The division of modules in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present application may be integrated into a processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. It is understood that the functions or implementations of each module in the embodiments of the present application may further refer to the relevant description of the method embodiment.

[0233] In one possible approach, the communication device may be as follows Fig.13As shown, the communication device may be a communication device or a chip in a communication device, wherein the communication device may be a network intelligence unit or an inference unit. The device may include a processor 1301, a communication interface 1302, and a memory 1303. Among them, the processing module 1202 may be the processor 1301. The communication module 1201 may be the communication interface 1302.

[0234] The processor 1301 may be a central processing unit (CPU), or a digital processing module, etc. The communication interface 1302 may be a transceiver, or an interface circuit such as a transceiver circuit, or a transceiver chip, etc. The device further includes: a memory 1303 for storing programs executed by the processor 1301. The memory 1303 may be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), etc., or a volatile memory (volatile memory), such as a random-access memory (RAM). The memory 1303 is any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0235] The processor 1301 is used to execute the program code stored in the memory 1303, specifically to execute the actions of the processing module 1202, which will not be described in detail in this application. The communication interface 1302 is specifically used to execute the actions of the communication module 1201, which will not be described in detail in this application.

[0236] The specific connection medium between the communication interface 1302, the processor 1301 and the memory 1303 is not limited in the embodiment of the present application. Fig.13 The memory 1303, the processor 1301 and the communication interface 1302 are connected via a bus 1304. Fig.13 The connections between the other components are shown in bold lines, which are only for illustration and are not intended to be limiting. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.13 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0237] An embodiment of the present invention further provides a computer-readable storage medium for storing computer software instructions required to be executed by the above-mentioned processor, which includes a program required to be executed by the above-mentioned processor.

[0238] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0239] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0240] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0241] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0242] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A network operation and maintenance method, characterized in that: The method comprises: The network intelligence unit obtains first network data; The network intelligence unit is trained based on the first network data to obtain a data model, wherein the data model is used to determine a network control instruction sent to a network element / network management system EMS / NMS, and the first network data comes from the EMS / NMS or a network element managed by the EMS / NMS; The network intelligence unit determines that the reasoning unit of the data model to be deployed is the reasoning unit included in the autonomous domain control unit, and the autonomous domain control unit is used to send a network control instruction to the EMS / NMS according to the reasoning result, or the autonomous domain control unit is deployed in the EMS / NMS; The network intelligence unit sends the data model to the reasoning unit included in the autonomous domain management and control unit.

2. A network operation and maintenance device, characterized in that: The device comprises: A communication module, used for acquiring first network data; a processing module, configured to perform training based on the first network data acquired by the communication module to obtain a data model, wherein the data model is used to determine a network control instruction sent to a network element / network management system EMS / NMS, and the first network data comes from the EMS / NMS or a network element managed by the EMS / NMS; And, determining that the inference unit of the data model to be deployed is an inference unit included in the autonomous domain control unit, the autonomous domain control unit is used to send a network control instruction to the EMS / NMS according to the inference result, or the autonomous domain control unit is deployed in the EMS / NMS; And, sending the data model to the reasoning unit included in the autonomous domain management and control unit through the communication module.

3. A network operation and maintenance system, characterized in that: The system comprises: the network operation and maintenance device according to claim 2 and a first network operation and maintenance device, wherein the first network operation and maintenance device comprises: A communication module, used for receiving a data model sent by the network operation and maintenance device; The processing module is used to: obtain second network data, where the second network data is the network data to be input into the data model for reasoning; input the second network data into the data model for reasoning to obtain a network control instruction; and execute the network control instruction.

Images