CN115442819A

CN115442819A - Network optimization method and communication device

Info

Publication number: CN115442819A
Application number: CN202110611317.9A
Authority: CN
Inventors: 李文璟; 丰雷; 喻鹏; 赵明宇; 严学强; 吴建军; 周凡钦; 高静
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2021-06-01
Filing date: 2021-06-01
Publication date: 2022-12-06

Abstract

The application provides a network optimization method and a communication device. The method comprises the following steps: the method comprises the steps that a network management node obtains first network operation data of a target network, and determines a first network problem based on the first network operation data and network problem judgment information in a knowledge graph; determining a first network optimization scheme for the first network problem based on network optimization decision logic in the knowledge graph, the network optimization decision logic being used for representing the optimization scheme for the network problem; and the network management node sends the information of the first network optimization scheme to a network element in the target network. The network problem judgment information comprises judgment logic of the network problem and network information, the network problem judgment logic is used for expressing judgment conditions of the network problem and the network problem corresponding to the judgment conditions, and the network information comprises network configuration resource information or network environment information. The first network optimization scheme is the most powerful network optimization scheme to solve the first network problem. The method can efficiently complete network optimization.

Description

Network optimization method and communication device

Technical Field

The present application relates to the field of communications technologies, and in particular, to a network optimization method and a communications apparatus.

Background

With the development of communication technology, various network problems may exist in a network, and in order to ensure the network performance of the network, the problem network needs to be optimized.

At present, a manual intervention with abundant experience is needed to accurately analyze network problems existing in a network and solutions corresponding to the network problems so as to complete network optimization, but the method cannot realize automation of network optimization.

Disclosure of Invention

The application provides a network optimization method and a communication device, which aim to realize the automation of network optimization.

In a first aspect, a method for network optimization is provided, where the method includes: the method comprises the steps that a network management node obtains first network operation data of a target network, wherein the first network operation data are used for reflecting the current operation state of the target network; the network management node determines a first network problem based on the first network operation data and the network problem judgment information in the knowledge graph, wherein the network problem judgment decision logic is used for representing the judgment condition of the network problem and the network problem corresponding to the judgment condition; the network management node determines a first network optimization scheme of the first network problem based on network optimization decision logic in the knowledge graph, wherein the network optimization decision logic is used for representing the optimization scheme of the network problem; and the network management node sends the first network optimization scheme to the network element in the target network.

According to the network optimization method and the network optimization system, the first network operation data of the target network are obtained through the network management node, the current network problems of the target network are determined according to the first network operation data and the network problem judgment logic in the knowledge graph, and the first network optimization scheme corresponding to the first network problems is determined based on the network optimization decision logic in the knowledge graph, so that network optimization can be completed by network elements in the target network based on the first network optimization scheme without depending on manual experience to determine the network problems and the network optimization scheme, labor cost and optimization time cost are effectively reduced, and timeliness of network optimization is improved.

The above manner of acquiring the first network operation data of the target network by the network management node may be directly acquired, for example, the network management node periodically collects the network operation data. The network management node may also indirectly obtain the first network operation data, for example, the network management node collects the user complaint data according to a requirement, and then obtains the first network operation data from the knowledge graph according to the collected user complaint data.

The network optimization scheme in the application can be understood as measures taken to solve network problems existing in the network. Illustratively, the network has a network problem that the power of the main cell is low, and a power control algorithm is adopted to increase the power of the main cell of the network, wherein the power control algorithm can be understood as a network optimization scheme.

In the embodiment of the present application, the network operation data before the network optimization is referred to as first network operation data, and the network operation data after the network optimization is referred to as second network operation data.

With reference to the first aspect, in some implementations of the first aspect, the network problem judgment information includes network problem judgment decision logic and network information, the network problem judgment logic is configured to indicate a judgment condition of a network problem and a network problem corresponding to the judgment condition, and the network information includes network configuration resource information or network environment information.

With reference to the first aspect, in certain implementations of the first aspect, after the network management node sends the first network optimization scheme to a network element in the target network, the method further includes: the network management node acquires second network operation data of the target network, wherein the second network operation data is operation data of the target network after optimization according to a first network optimization scheme; the network management node judges whether the second network operation data meets the network operation quality evaluation standard or not; and if the second network operation data does not accord with the network operation quality evaluation standard, determining whether a second network optimization scheme of the first network problem exists, wherein the network operation data after the target network is optimized according to the second network optimization scheme accords with the network operation quality evaluation standard.

Illustratively, the above-mentioned network operation quality evaluation criterion may be understood as a threshold value set in advance. Namely, whether the second network operation data reaches the threshold value is judged, if so, the optimization requirement is met, and the optimization task is ended. If the network optimization scheme does not reach the threshold value, the network optimization scheme of the network problem is determined again, the optimization task is continued until a network optimization scheme (namely the second network optimization scheme) exists, so that the second network operation data can reach the threshold value, and the optimization task is ended.

In the embodiment of the application, the network optimization scheme corresponding to the second network operation data meeting the network operation quality evaluation standard is referred to as a second network optimization scheme.

With reference to the first aspect, in certain implementations of the first aspect, the first network optimization scheme is a scheme with the highest power among a plurality of network optimization schemes for solving the first network problem.

With reference to the first aspect, in some implementations of the first aspect, the determining whether the second network optimization scheme exists includes: and the network management node sequentially performs network optimization according to the sequence of the success rate of the plurality of network optimization schemes for solving the first network problem from high to low until the second network optimization scheme exists.

It should be understood that the above-mentioned network optimization decision logic may also represent the success rate of the optimization scheme of the network problem on the basis of the optimization scheme of the network problem, and the success rate of the optimization scheme represents the success rate of solving the network problem by using the optimization scheme. Illustratively, the success rate of the optimization scheme of the network problem may be a success rate determined based on experience of historical network optimization.

With reference to the first aspect, in certain implementations of the first aspect, the method further includes: and if the second network optimization scheme does not exist, re-determining the network problem of the target network.

When the network problem existing in the target network is determined again, the network problem with the second highest accuracy can be selected, and a network optimization scheme for solving the network problem is determined in the network optimization decision logic according to the network problem selected again.

With reference to the first aspect, in some implementations of the first aspect, the first network question is a network question with a highest accuracy among a plurality of network questions determined according to the network question determination decision logic.

With reference to the first aspect, in certain implementations of the first aspect, the above re-determining a network problem existing in the target network includes: and the network management node sequentially determines the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists and the network operation data of the target network optimized according to the target network optimization scheme meets the network operation quality evaluation standard.

It should be appreciated that the above-described network problem determination decision logic is also used to represent the accuracy of the network problem, which represents the accuracy of the network problem determined from the determination conditions of the network problem. Illustratively, the accuracy of the network problem may be an accuracy determined based on experience with historical network optimization.

With reference to the first aspect, in certain implementations of the first aspect, the method further includes: and under the condition that the second network operation data does not meet the network operation quality evaluation standard, updating the network optimization decision logic by taking the success rate of the second network optimization scheme as the highest-power scheme in a plurality of network optimization schemes for solving the first network problem, and/or updating the network judgment logic by taking the accuracy rate of the target network problem as the highest-accuracy network problem in the network problem judgment decision logic.

It should be understood that, after network optimization is sequentially performed on network elements in the target network according to an optimization scheme corresponding to network problems existing in the knowledge graph, under the condition that the obtained network operation data do not meet the network operation quality evaluation standard, the network problems existing in the target network and corresponding solutions can be re-determined by a manual intervention method, the determined network problems are stored in the knowledge graph, the network problem logic is updated, the solutions corresponding to the determined network problems are stored in the knowledge graph, and the network optimization decision logic is updated.

With reference to the first aspect, in some implementations of the first aspect, the acquiring, by the network management node, first network operation data of the target network includes: the network management node carries out natural language processing on the complaint data of the user to obtain objective intention information of the user, wherein the complaint data of the user is used for reflecting the network state of the target network; the network management node determines basic resource data of a target network based on the objective intention information of the user; and the network management node acquires the first network operation data of the target network from the network information in the knowledge graph based on the basic resource data.

It should be understood that the complaint data of the user is a subjective expression of the user's intention, and the network management node may perform natural language processing on the subjective expression of the user to obtain objective intention information that can be recognized by the machine, and may be represented in the following form:

[ entity, phenomenon; action, object attribute

It should also be understood that the basic resource data includes information of access network equipment where the user is located, information of a cell, or location information of a base station, etc.

According to the embodiment of the application, the objective intention of the user is introduced, the representation form of the objective intention of the user is provided, and the translation from the subjective intention of the user to the objective intention is realized by adopting natural language processing, so that the machine can identify the intention of the user, problems are basically not needed to be analyzed by human intervention with abundant experience, the labor cost and the optimization time cost are reduced, and the network optimization efficiency is improved.

In a second aspect, a method for constructing a knowledge graph is provided, the method comprising: the data processing equipment acquires network information and stores the network information into the knowledge graph in a knowledge form to obtain the network information in the knowledge graph; the data processing equipment determines a network problem judgment decision logic based on historical operation data of a target network and the network problems, and stores node information in the network problem judgment decision logic into the knowledge map in a knowledge form, wherein the network problem judgment decision logic is used for representing judgment conditions of the network problems and the network problems corresponding to the judgment conditions.

With reference to the second aspect, in some implementations of the second aspect, the data processing device determines a network optimization decision logic based on the historical network problems and the optimization solutions of the network, and stores node information in the network optimization decision logic in a knowledge form in the knowledge graph, where the network optimization decision logic is used to represent the optimization solutions of the network problems.

With reference to the second aspect, in some implementations of the second aspect, the data processing device obtains a network operation quality evaluation criterion according to historical operation data of the network, and stores the network operation quality evaluation criterion in the knowledge graph in a knowledge form to obtain the network operation quality evaluation criterion in the knowledge graph.

In a third aspect, a communication apparatus is provided, including: the system comprises a transceiving module, a first network operation module and a second network operation module, wherein the transceiving module is used for acquiring first network operation data of a target network, and the first network operation data is used for reflecting the current operation state of the target network; the processing module is used for determining a first network problem based on the first network operation data and network problem judgment decision logic in the knowledge graph, wherein the network problem judgment decision logic is used for representing a judgment condition of the network problem and the network problem corresponding to the judgment condition; and determining a first network optimization scenario for the first network problem based on network optimization decision logic in the knowledge-graph, the network optimization decision logic being for representing an optimization scenario for a network problem; the transceiver module is further configured to: and sending the first network optimization scheme to a network element in the target network.

Optionally, the network problem judgment information includes a network problem judgment decision logic and network information, the network problem judgment logic is configured to indicate a judgment condition of a network problem and a network problem corresponding to the judgment condition, and the network information includes network configuration resource information or network environment information.

Optionally, the transceiver module is further configured to: acquiring second network operation data of the target network, wherein the second network operation data is operation data of the target network after optimization according to the first network optimization scheme; the processing module is further configured to: judging whether the second network operation data meet a network operation quality evaluation standard or not; and if the second network operation data does not accord with the network operation quality evaluation standard, determining whether a second network optimization scheme of the first network problem exists, wherein the network operation data optimized by the target network according to the second network optimization scheme accords with the network operation quality evaluation standard.

Optionally, the first network optimization scheme is a scheme with the highest power among a plurality of network optimization schemes for solving the first network problem.

Optionally, the processing module is further configured to: and sequentially performing network optimization according to the sequence of the success rate of the plurality of network optimization schemes for solving the first network problem from large to small until the second network optimization scheme exists.

Optionally, the processing module is further configured to: and if the second network optimization scheme does not exist, re-determining the network problem of the target network.

Optionally, the first network problem is a network problem with the highest accuracy among the plurality of network problems determined according to the network problem judgment decision logic.

Optionally, the processing module is further configured to: and sequentially determining the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists, wherein the network operation data of the target network after being optimized according to the target network optimization scheme meets the network operation quality evaluation standard.

Optionally, the processing module is further configured to: and under the condition that the second network operation data does not meet the network operation quality evaluation standard, updating the network optimization decision logic by taking the second network optimization scheme as the most powerful scheme in a plurality of network optimization schemes for solving the first network problem.

Optionally, the processing module is further configured to: and updating the network problem judgment decision logic by taking the accuracy of the target network problem as the network problem with the highest accuracy in the network problem judgment decision logic.

Optionally, the processing module is further configured to: natural language processing is carried out on complaint data of a user to obtain objective intention information of the user, wherein the complaint data of the user is used for reflecting the network state of the target network; and determining basic resource data of the target network based on the objective intention information of the user; the transceiver module is further configured to: and acquiring first network operation data of the target network from the network information in the knowledge graph based on the basic resource data.

In a fourth aspect, a network optimization device is provided, which includes a processor coupled with a memory and configured to execute instructions in the memory to implement the method in any one of the possible implementation manners of the first aspect and/or the second aspect. Optionally, the apparatus further comprises a memory. Optionally, the apparatus further comprises a communication interface, the processor being coupled to the communication interface.

In one implementation, the network optimization device is a network management node. When the network optimization device is a network management node, the communication interface may be a transceiver, or an input/output interface.

In a fifth aspect, a processor is provided, which includes: input circuit, output circuit and processing circuit. The processing circuit is configured to receive a signal via the input circuit and transmit a signal via the output circuit, so that the processor performs the method of any of the possible implementations of the first aspect and/or the second aspect.

In a specific implementation process, the processor may be a chip, the input circuit may be an input pin, the output circuit may be an output pin, and the processing circuit may be a transistor, a gate circuit, a flip-flop, various logic circuits, and the like. The input signal received by the input circuit may be received and input by, for example and without limitation, a receiver, the signal output by the output circuit may be output to and transmitted by a transmitter, for example and without limitation, and the input circuit and the output circuit may be the same circuit that functions as the input circuit and the output circuit, respectively, at different times. The embodiment of the present application does not limit the specific implementation manner of the processor and various circuits.

In a sixth aspect, a processing apparatus is provided that includes a processor and a memory. The processor is configured to read instructions stored in the memory, and may receive a signal via the receiver and transmit a signal via the transmitter to perform the method of any one of the possible implementations of the first aspect and/or the second aspect.

Optionally, there are one or more processors and one or more memories.

Alternatively, the memory may be integrated with the processor, or provided separately from the processor.

In a specific implementation process, the memory may be a non-transient memory, such as a Read Only Memory (ROM), which may be integrated on the same chip as the processor, or may be separately disposed on different chips.

It will be appreciated that the associated data interaction process, for example, sending the indication information, may be a process of outputting the indication information from the processor, and receiving the capability information may be a process of receiving the input capability information from the processor. In particular, the data output by the processor may be output to a transmitter and the input data received by the processor may be from a receiver. The transmitter and receiver may be collectively referred to as a transceiver, among others.

The processing means in the above fifth aspect may be a chip, the processor may be implemented by hardware or may be implemented by software, and when implemented by hardware, the processor may be a logic circuit, an integrated circuit, or the like; when implemented in software, the processor may be a general-purpose processor implemented by reading software code stored in a memory, which may be integrated with the processor, located external to the processor, or stand-alone.

In a seventh aspect, a computer program product is provided, the computer program product comprising: computer program (also called code, or instructions), which when executed, causes a computer to perform the method of any of the possible implementations of the first aspect and/or the second aspect.

In an eighth aspect, a computer-readable storage medium is provided, which stores a computer program (which may also be referred to as code or instructions) that, when executed on a computer, causes the computer to perform the method of any one of the possible implementations of the first aspect and/or the second aspect.

Drawings

FIG. 1 is a schematic block diagram of a system architecture provided by an embodiment of the present application;

FIG. 2 is a schematic illustration of graph data in a knowledge graph provided by an embodiment of the application;

fig. 3 is a schematic flow chart of a network optimization method provided in an embodiment of the present application;

FIG. 4 is a schematic flow chart diagram of a method for constructing a knowledge graph provided by an embodiment of the application;

fig. 5 is a schematic flow chart of another network optimization method provided in the embodiments of the present application;

FIG. 6 is a schematic flow chart diagram of a knowledge-graph updating method provided by an embodiment of the application;

fig. 7 is a schematic flow chart of another network optimization method provided in the embodiments of the present application;

FIG. 8 is a schematic diagram of a network problem determination decision tree according to an embodiment of the present application;

FIG. 9 is a schematic diagram of a network optimization scheme decision tree provided by an embodiment of the present application;

fig. 10 is a schematic block diagram of a communication device provided in an embodiment of the present application;

fig. 11 is a schematic block diagram of a network optimization device according to an embodiment of the present application.

Detailed Description

The technical solution in the present application will be described below with reference to the accompanying drawings.

A system architecture suitable for use in embodiments of the present application is described in detail below with reference to fig. 1.

Fig. 1 illustrates a system architecture 100 provided by the present application. As shown in fig. 1, the system architecture 100 includes a network management node 110, an access network 120, and a core network 180.

It should be understood that the network management node 110 may be deployed in a centralized manner or in a distributed manner, which is not limited in this application.

It should be understood that the network management node 110 described above is used for network management. For example, the network management node may be a radio access network element management system (OMC-R), an apparatus having an Operation Support System (OSS), an apparatus having a network data analysis function (NWDAF), or a virtual core network apparatus.

The access network 120 may include an evolved Node B (eNB or eNodeB) in a Long Term Evolution (LTE) system, a home base station (e.g., home evolved Node B or home Node B, HNB), a Base Band Unit (BBU), or a macro base station (e.g., macro eNB or macro nb), a base station corresponding to a small cell (small cell), where the small cell may include: an urban cell (metro cell), a micro cell (pico cell), a femto cell (femto cell), or the like, or a gNB in a New Radio (NR) system, a satellite base station in a satellite communication system, or the like, or a RAN device including a RAN device of a CU node and a DU node, a control plane CU node (CU-CP node), a user plane CU node (CU-UP node), and a DU node, which is not limited in the embodiment of the present application.

The core network 180 may be a 5G core network or a 4G core network. Wherein, the 5G core network element includes: access and mobility management function (AMF), session Management Function (SMF), user Plane Function (UPF), network Exposure Function (NEF), network warehousing function (NRF), policy Control Function (PCF), network Slice Selection Function (NSSF), unified data management function (UDM), authentication service function (AUSF). The 4G core network element comprises: a Home Subscriber Server (HSS), a Mobile Management Entity (MME), a Policy and Charging Rules Function (PCRF), a serving gateway (S-GW), and a public data network gateway (PGW).

It should be understood that besides the devices shown in fig. 1, a wireless controller in a Cloud Radio Access Network (CRAN) scenario may be included, or a relay station, an access point, a vehicle-mounted device, a wearable device, and a network device in a 5G network or a network device in a PLMN network that is evolved in the future, or an Access Point (AP) in a WLAN.

It should also be understood that fig. 1 is a simplified schematic diagram that is shown only for ease of understanding, and that other network elements, not shown in fig. 1, may also be included in the system architecture 100.

To facilitate understanding of the embodiments of the present application, related terms referred to in the present application will be first introduced.

Knowledge graph: refers to a knowledge base that integrates data using a graphical structured data model or topology. Compared with the traditional database, the graphical structured data model or the topological structure in the knowledge graph provides a flexible design mode, and labels or attributes can be added to the original data source only by adding a new mode definition. The knowledge graph includes a graph describing various entities or concepts and their interrelations existing in the real world, the entities or concepts or values are represented by nodes, and the edges represent attributes or relationships between the nodes. As shown in fig. 2, a semantic network graph is formed by nodes and edges, knowledge resources and carriers thereof are described by using a visualization technology, and knowledge and interrelations between the knowledge resources and the carriers are mined, analyzed, constructed, drawn and displayed. The network management node can query the edges (representing the relation with the opposite terminal entity) of various entities in the knowledge graph, thereby rapidly acquiring the opposite terminal entity connected with the network management node and realizing high-efficiency query.

In the embodiment of the present application, in the data acquisition stage, the network information knowledge graph may be constructed through a data model in the form of, for example, a triplet (radio access network, whether or not it includes access network equipment), (cell 1, adjacency, cell 2), or the like. In the network problem determination stage, a network problem determination knowledge map is formed by means of a form such as a triplet (network problem determination condition, network problem classification probability, determined network problem). In the network optimization stage, a network optimization solution knowledge graph is formed through data models in the form of triples (judged network problem, whether the problem can be called or not, and optimization solution for the problem). The different stages form the entire network knowledge graph.

Exemplary knowledge in the knowledge-graph may include: the system comprises user subjective corpus information, user objective intention information, network configuration resource information, network environment information (such as geographic environment, climate environment, living environment, electromagnetic environment and the like), network operation quality judgment information, network optimization problem judgment rules, network optimization scheme making rules and the like, wherein the information or the rules are stored in a knowledge map in a knowledge mode and can be dynamically updated according to changes of actual environment and evaluation of each network optimization effect.

It should be understood that the knowledge graph may be placed on a server alone or in a network management system.

In the early stage of network optimization, a network optimization scheme is usually formulated through analysis of OMC-R data and drive test data. However, it is difficult to find and solve problems only by this method, and at this time, problems are often found by combining with user complaints and CQT test methods, and the root causes of the problems are analyzed and found by combining with signaling tracking analysis, traffic statistical analysis, drive test analysis, and the like, so that an optimization scheme can be made in a targeted manner.

Although the method can accurately position the problems existing in the network, the network optimization is basically problem-driven, namely, the network optimization is triggered after the problems (such as user complaints, rapid degradation of network performance, network alarm and the like) occur, and a large amount of time, manpower and material resources are consumed in the process of network optimization. Particularly, when a network problem is obtained based on a method of user complaint, because the complaint user is not a professional, the description problem may deviate from the actual problem, which brings difficulty to the formulation of a network optimization scheme, requires manual intervention with abundant experience to accurately analyze the problem, and has poor automation and timeliness.

The application provides a network optimization method and a communication device for improving network optimization efficiency.

Before describing the network optimization method and the communication device provided in the embodiments of the present application, the following description is made.

First, in the embodiments shown below, terms and english abbreviations such as network problem judgment decision logic, network optimization decision logic, etc. are exemplary examples given for convenience of description, and should not limit the present application in any way. This application is not intended to exclude the possibility that other terms may exist or may be defined in the future that achieve the same or similar functionality.

Second, the first, second and various numerical numbering in the embodiments shown below are merely for convenience of description and are not intended to limit the scope of the embodiments of the present application. For example, to distinguish between different network operating data, to distinguish between different network optimization schemes, etc. It should be noted that the ordinal numbers such as "first", "second", etc. are used in the embodiments of the present application to distinguish a plurality of objects, and are not used to limit the sequence, timing, priority or importance of the plurality of objects. For example, the first network operation data, the second network operation data, and the like are only used for distinguishing the network operation data before and after different optimizations, and do not indicate that the two network operation data have different structures, different degrees of importance, and the like.

Third, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, and c, may represent: a, or b, or c, or a and b, or a and c, or b and c, or a, b and c, wherein a, b and c can be single or multiple.

The method 300 for network optimization according to the embodiment of the present application is described in detail below with reference to fig. 3. The method 300 may be applied to the system architecture 100 shown in fig. 1, and may also be applied to other system architectures, but the embodiment of the present application is not limited thereto.

It should be understood that the knowledge graph in the embodiments of the present application may be deployed in a network management node.

Fig. 3 is a schematic flow chart of a network optimization method 300 provided in an embodiment of the present application. As shown in fig. 3, the method 300 includes the steps of:

s301, the network management node obtains first network operation data of the target network, wherein the first network operation data is used for reflecting the current operation state of the target network.

It should be understood that the above manner in which the network management node acquires the first network operation data of the target network may be directly acquired, for example, the network management node periodically collects the network operation data. The network management node may also indirectly obtain the first network operation data. For example, the network management node collects the user complaint data according to the requirements, and then acquires the first network operation data from the knowledge graph according to the collected user complaint data.

S302, the network management node determines a first network problem based on the first network operation data and the network problem judgment information in the knowledge graph, and the network problem judgment decision logic is used for representing the judgment condition of the network problem and the network problem corresponding to the judgment condition.

The network problem judgment information in the embodiment of the application comprises network problem judgment decision logic and network information. The network information may specifically include: network configuration resource information or network environment information.

The network configuration resource information refers to configuration data of an access network, a core network, a cloud data center and various terminals.

Network environment information refers to the sum of information including objective things surrounding an entity (people, access network equipment, servers, vehicles, etc.). Generally, the information is divided into natural environment information and living environment information.

The network configuration resource information may include a terminal, a network element, a link and a relationship therebetween related to the target network, and the network environment information may include geographic environment information, climate environment information, living environment information, and the like where the terminal or the network element is located. For example, the relationship between the network elements may be a neighboring relationship (adjacent or non-adjacent) between cells, and the environment information may be information such as the density of buildings around the target network element.

After the network management node obtains the first network operation data, the network management node may determine which type of network element the network problem existing in the target network belongs to (for example, the network problem may be access network equipment, an exchange, or other types of network elements) according to the network problem determination decision logic, and then determine which base station or which exchange the network problem specifically belongs to (that is, the specific location where the network element generating the network problem is located) based on the network information in the network problem determination decision information.

S303, the network management node determines a first network optimization scheme of the first network problem based on a network optimization decision logic in the knowledge graph, where the network optimization decision logic is used to represent an optimization scheme of the network problem.

The network optimization decision information comprises one or more network problems and one or more optimization schemes corresponding to the network problems. The network optimization scheme in the application can be understood as measures taken to solve network problems existing in the network. Illustratively, the network has a network problem that the power of the main cell is low, and a power control algorithm is adopted to increase the power of the main cell of the network, wherein the power control algorithm can be understood as a network optimization scheme.

S304, the network management node sends the first network optimization information to the network element in the target network.

And the network element completes network optimization according to the received first network optimization scheme. In the embodiment of the present application, the network operation data before the network optimization is referred to as first network operation data, and the network operation data after the network optimization is referred to as second network operation data.

According to the method and the device, the first network operation data of the target network are obtained, the decision logic is judged according to the first network operation data and the network problems in the knowledge graph to determine the current network problems of the target network, and then the first network optimization scheme corresponding to the first network problems is determined based on the network optimization decision logic in the knowledge graph. Therefore, the network element in the target network can complete network optimization based on the first network optimization scheme, the problem that the network optimization scheme is determined by depending on manual experience is effectively solved, labor cost and optimization time cost are reduced, and network optimization efficiency is improved.

As an optional embodiment, after the network management node sends the first network optimization scheme to the network element in the target network, the method 300 further includes:

s305, the network management node acquires the second network operation data of the target network. The second network operation data is the operation data after the network element in the target network is optimized according to the first network optimization scheme. The network management node judges whether the second network operation data meets the network operation quality evaluation standard. And if the second network operation data does not accord with the network operation quality evaluation standard, determining whether a second network optimization scheme of the first network problem exists or not, so that the network operation data after the network element in the target network is optimized according to the second network optimization scheme accords with the network operation quality evaluation standard.

It should be understood that the above-mentioned network operation quality evaluation criterion is used to determine whether the relevant index of the operation quality of the current network meets the threshold value preset by the relevant index. For example, the network operation quality evaluation criterion may be measured by some indicators. For example, reference Signal Received Power (RSRP), channel Quality Indicator (CQI), received Signal Strength Indicator (RSSI), or the like.

The RSRP is an important index for measuring the coverage of the wireless network of the system. In the actual judgment process, the network management node presets the range value of RSRP, and if the RSRP does not belong to the range, the currently judged network operation quality does not reach the standard. For example, in the LTE system, the terminal generally reports RSRP in the range of [ -140 decibel-milliwatt (dBm), -44dBm ], that is, the RSRP is out of the range, which indicates that the operation quality of the network does not reach the standard. When the road is measured on dense urban areas, general urban areas and key traffic main lines, the RSRP is generally required to be larger than-100 dBm, otherwise the problems of call drop, weak coverage and the like are easy to occur.

The CQI is an important index for measuring the quality of wireless channel communication, and represents the quality of the current channel, which corresponds to the signal-to-noise ratio of the channel. The common value of the CQI is 6-12, and when the value of the CQI exceeds the value range of the indexes, the current network operation quality does not reach the standard.

RSSI is used to determine the link quality and whether to increase the broadcast transmission strength. The normal range of the RSSI reported by the terminal is [ -90dBm, -25dBm ], and if the normal range exceeds the range, the RSSI is considered to be abnormal, namely the current network operation quality does not reach the standard. It should be appreciated that RSSI values outside the above range can have significant impact on call quality, dropped calls, congestion, network coverage, capacity, etc.

It should be understood that there may be other ways to evaluate the network operation quality criteria for network optimization, which is not limited in this application.

The above-mentioned network operation quality evaluation criterion may be understood as a threshold value set in advance. Namely, whether the second network operation data reaches the threshold value is judged, if the second network operation data reaches the threshold value, the optimization requirement is met, and the optimization task is finished. If the network optimization scheme does not reach the threshold value, the network optimization scheme of the network problem is determined again, the optimization task is continued until a network optimization scheme (namely the second network optimization scheme) exists, so that the second network operation data can reach the threshold value, and the optimization task is ended.

It should also be appreciated that the first network problem determined at S302 above may correspond to at least one network optimization scenario in the knowledge-graph. Under the condition that one network problem corresponds to a plurality of network optimization schemes, the network management node selects any network optimization scheme capable of solving the network problem to carry out network optimization. If the target network completes network optimization according to any network optimization scheme selected by the network management node, and the obtained second network operation data does not accord with the network operation quality evaluation standard, the network management node needs to re-determine another network optimization scheme corresponding to the network problem, and enable the network elements in the target network to perform network optimization according to the another network optimization scheme until the second network optimization scheme exists, so that the network operation data after the network elements in the target network perform optimization according to the second network optimization scheme accords with the network operation quality evaluation standard.

According to the embodiment of the application, the network operation quality evaluation standard is set, the result of network optimization can be judged more conveniently, the result of network optimization does not need to be judged manually, the labor cost is effectively reduced, and the autonomy and the optimization efficiency of network optimization are further improved.

As an optional embodiment, the first network optimization scheme is a scheme with the highest power among a plurality of network optimization schemes for solving the first network problem. The determining whether the second network optimization scheme exists includes: and the network management node sequentially performs network optimization according to the sequence of the success rate of solving the first network problem by the plurality of network optimization schemes from large to small until the second network optimization scheme exists.

After the first network problem is determined in S302, a network optimization scheme with the highest success rate for solving the network problem is generally selected for network optimization. If the network operation data optimized by the optimization scheme with the highest success rate does not meet the network operation quality evaluation standard, the network management node may sequentially select the corresponding network optimization schemes according to the sequence from the highest success rate to the lowest success rate to perform network optimization until the second network operation data after the network element in the target network performs network optimization according to the network optimization scheme selected by the network management node meets the network operation quality evaluation standard.

Illustratively, there are 4 network optimization schemes that can solve the first network problem in the knowledge-graph, with success rates of 0.6, 0.3, 0.15, and 0.05, respectively. After the network management node determines the first network problem, a network optimization scheme with the highest success rate (namely 0.6) is selected for network optimization, and if the second network operation data obtained after optimization does not meet the network operation quality evaluation standard, the network management node can select a network optimization scheme with the highest success rate (namely 0.3) for network optimization. And if the optimization result still does not accord with the network operation quality evaluation standard, selecting a network optimization scheme with the success rate of 0.15 for network optimization. If the operation data of the network element in the target network after network optimization according to the network optimization scheme with the success rate of 0.15 meets the network operation quality evaluation standard, the optimization is finished, and the first network problem is successfully solved.

As an alternative embodiment, the method 300 further comprises: and if the second network optimization scheme does not exist, re-determining the network problem of the target network.

With reference to the above example, if the operation data obtained after the network element in the target network performs network optimization according to the network optimization scheme with the success rate of 0.05 does not still meet the network operation quality evaluation standard, it is determined that the first network problem determined in S302 is incorrect, and the network problem existing in the target network is determined again.

As an alternative embodiment, the first network problem is a network problem with the highest accuracy among the multiple network problems determined by the network problem determination decision logic. The above network problem of re-determining the existence of the target network includes: and the network management node sequentially determines the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists and network operation data of a network element in the target network after the network element is optimized according to the target network optimization scheme meets the network operation quality evaluation standard.

The determination condition for each network issue in the knowledge-graph may correspond to at least one network issue. In the case where one network problem determination condition corresponds to a plurality of network problems, the network management node generally determines the network problem with the highest accuracy as the first network problem. If the second network operation data after the network element in the target network is optimized according to the network optimization scheme of the first problem does not meet the network operation quality evaluation standard, the network management node can sequentially determine the network problems according to the sequence of the accuracy rate of the network problems from large to small, and select a plurality of network optimization schemes capable of solving the network problems according to the determined network problems until a network optimization scheme of the network problems exists, so that the network operation data after the network element in the target network is optimized according to the optimization scheme meets the network operation quality evaluation standard.

Illustratively, the judgment condition of a certain network problem in the knowledge graph corresponds to 3 network problems, and the accuracy rates of the network problems are 0.7, 0.15 and 0.1 respectively. The network optimization schemes capable of solving the network problem with the accuracy rate of 0.7 in the knowledge graph are 3, the network optimization schemes capable of solving the network problem with the accuracy rate of 0.15 are 2, and the network optimization schemes capable of solving the network problem with the accuracy rate of 0.1 are 4. In this case, the network management node may determine the network problem with the accuracy of 0.7 as the first network problem, and sequentially select 3 network optimization schemes that can solve the first network problem according to the sequence from high to low success rate, so as to perform network optimization. And assuming that the optimized network operation data do not accord with the network operation quality evaluation standard, the network management node determines the network problem with the accuracy rate of 0.15 as a second network problem, and sequentially selects 2 network optimization schemes capable of solving the second network problem according to the sequence of the success rate from high to low to perform network optimization. And if the optimized network operation data meets the network operation quality evaluation standard according to the 2 nd network optimization scheme of the second network problem, the second network problem is the target network problem, and the 2 nd network optimization scheme of the second network problem is the target network optimization scheme. If the optimized network operation data does not accord with the network operation quality evaluation standard according to the 2 network optimization schemes of the second network problem, the network management node determines the network problem with the accuracy rate of 0.1 as a third network problem, and sequentially selects 4 network optimization schemes capable of solving the third network problem according to the sequence from high success rate to low success rate to perform network optimization until a target network optimization scheme of the target network problem exists.

In the embodiment of the present application, when the network problem existing in the target network is determined again, the network problem with the second highest accuracy may be selected, and a network optimization scheme for solving the network problem is determined in the network optimization decision logic according to the network problem selected again, and the specific process may refer to the process of finding the second network optimization scheme, which is not described herein again.

As an alternative embodiment, the method 300 further comprises: and under the condition that the second network operation data does not meet the network operation quality evaluation standard, updating the network optimization decision logic by taking the second network optimization scheme as the scheme with the highest power in a plurality of network optimization schemes for solving the first network problem.

Combining with the examples that the number of network optimization schemes capable of solving the first network problem in the knowledge graph is 4 and the success rates are 0.6, 0.3, 0.15 and 0.05, the network management node determines the network optimization scheme with the success rate of 0.6 as the first network optimization scheme, and when network optimization is performed, the obtained second network operation data does not meet the network operation quality evaluation standard. And the network management node determines the network optimization scheme with the success rate of 0.3 as a second network optimization scheme again, and when network optimization is carried out, the obtained second network operation data meet the network operation quality evaluation standard, the success rate of 0.3 of the second network optimization scheme (the network optimization scheme with the success rate of 0.3) is updated to 0.6 so as to update the network optimization decision logic.

As an alternative embodiment, the method 300 further comprises: and under the condition that the second network operation data do not accord with the network operation quality evaluation standard, updating the network problem judgment decision logic by taking the target network problem as the network problem with the highest accuracy in the network problem judgment decision logic.

And in combination with the example that the judgment condition of a certain network problem in the knowledge graph corresponds to 3 network problems and the accuracy rates of the network problems are 0.7, 0.15 and 0.1 respectively, the network management node determines the network problem with the accuracy rate of 0.7 as a first network problem, and when network optimization is carried out, the obtained second network operation data does not accord with the network operation quality evaluation standard. And the network management node determines the network problem with the accuracy of 0.15 as a second network problem (namely the target network problem), and when network optimization is carried out, the obtained second network operation data meet the network operation quality evaluation standard, the accuracy of 0.15 of the target network (the network problem with the accuracy of 0.15) is updated to 0.7 so as to update the network problem judgment decision logic.

It should be understood that, after network optimization is sequentially performed by network elements in the target network according to the optimization schemes corresponding to the network problems existing in the knowledge graph, under the condition that the obtained network operation data do not meet the network operation quality evaluation standard, the network problems existing in the target network and the corresponding solutions can be re-determined by a manual intervention method, the determined network problems are stored in the knowledge graph, the network problem logic is updated, the solutions corresponding to the determined network problems are stored in the knowledge graph, and the network optimization decision logic is updated.

The network problem judgment decision logic and the network optimization decision logic in the knowledge graph can be dynamically updated, so that the information in the knowledge graph has timeliness and dynamics, and meanwhile, the accuracy of network optimization is improved.

As an optional embodiment, the acquiring, by the network management node, the first network operation data of the target network includes: the network management node processes the complaint data of the user to obtain objective intention information of the user, wherein the complaint data of the user is used for reflecting the network state of a target network; the network management node determines basic resource data of a target network based on the objective intention information of the user; the network management node obtains first network operation data of the target network from the network information in the knowledge graph based on the basic resource data.

[ entity, phenomenon; action, object attribute

Illustratively, the complaint information of user a is: "the net speed is extremely poor, how is not solved! ", the objective intent translated into: [ net speed, poor ], [ anger ]. The complaint information of the user B is: "what should be if the net is often unavailable," translates to an objective intent: [ net speed, difference ], [ dissatisfaction ].

It should also be understood that the above-mentioned basic resource data refers to the relevant resource data required by the user to access the communication network. For example, the basic resource data may include information of access network equipment where the user is located, information of a cell, or location information of a cell, and the like.

The knowledge graph comprises network operation data related to target network elements, namely peripheral network elements, after the basic resource data are obtained, the target network elements to which the complaint data of the user belong can be determined, and after the target network elements are determined, the network operation data of the target network elements are obtained from the knowledge graph.

Before the above network optimization, the present application also includes the following four alternative embodiments. The following four alternative embodiments provided herein may be performed by a data processing apparatus deploying a knowledge-graph generation tool.

As an optional embodiment, before the network management node obtains the first network operation data of the target network, the method further includes: and the data processing equipment acquires network information and stores the network information into the knowledge graph in a knowledge form to obtain the network information in the knowledge graph.

It should be understood that the network information in the knowledge-graph further includes: user intention information.

The user intention information comprises user subjective corpus information, user objective intention information, user psychological information and the like.

The embodiment of the present application may store the above network information in the form of a triplet, for example, (entity 1, relationship, entity 2) or (entity, attribute value). Or the network information is stored in an N-tuple form, for example, time, space, or other dimension information may be added on the basis of the triplet form, where N in this embodiment is an integer greater than 3.

It should be understood that the knowledge form in the embodiment of the present application may be in the form of the above-mentioned triplets or N-tuples, and may also be in other forms, for example, a relational database. This is not a limitation of the present application.

As an optional embodiment, before the network management node obtains the first network operation data of the target network, the method further includes: the data processing equipment obtains a network operation quality evaluation standard based on historical experience, and stores the network operation quality evaluation standard in a knowledge mode into a knowledge graph to obtain the network operation quality evaluation standard in the knowledge graph.

It should be understood that the historical experience may be an experience that a network optimization expert summarizes generalizations when performing network optimization at ordinary times.

It should also be appreciated that the above-described forms of knowledge may be stored in the knowledge-graph in the form of triples, e.g., (conditions of parameters or operating conditions, quality assessment indicators, quality assessment values).

As an optional embodiment, before the network management node obtains the first network operation data of the target network, the method further includes: the data processing equipment determines a network problem judgment decision logic based on historical operation data of a target network, and stores information in the network problem judgment decision logic into a knowledge graph in a knowledge form to obtain the network problem judgment decision logic in the knowledge graph, wherein the information in the network problem judgment decision logic represents judgment conditions of network problems and network problems corresponding to the judgment conditions.

The embodiment of the application stores the information in the network problem judgment decision logic in the knowledge graph in the form of triples or N-tuples. For example, the triplet form is (a set of value conditions of parameters or states, a network problem classification probability, a network problem classification), where examples of the network problem classification are as follows: the problem of equipment failure, the problem of weak coverage, the problem of uneven uplink and downlink coverage, the problem of cross-zone coverage, the problem of overlapping coverage, the problem of blocking interference, the problem of adjacent cell interference, the problem of additive noise interference, the problem of parameter configuration errors, the problem of switching and the like.

As an optional embodiment, before the network management node obtains the first network operation data of the target network, the method further includes: the data processing equipment obtains a network optimization decision logic based on historical network problems and an optimization scheme of a target network, and stores information in the network optimization decision logic into the knowledge graph in a knowledge form to obtain the network optimization decision logic in the knowledge graph, wherein the information in the network optimization decision logic represents the optimization scheme of the network problems.

It should be understood that the information in the network optimization decision logic is: and under the condition of a specific network problem classification given by the network problem judgment decision logic, giving a current network parameter or operation state, a network optimization category to be executed and an optimization algorithm to be called. Illustratively, the network optimization categories include: the method comprises the following steps of single station coverage optimization, parcel coverage optimization, indoor and outdoor collaborative optimization, adjacent cell optimization, handover parameter optimization, paging performance optimization, interference optimization, subway tunnel optimization, stadium optimization, intra-system load balancing optimization, inter-system load balancing optimization, or whole-network wireless index optimization.

The embodiment of the application stores the information in the network optimization decision logic in the knowledge graph in the form of triples or N-tuples. For example, the form of a quadruple is (classification of network problems, current network parameters or operating states, categories of network optimization to be performed, optimization algorithms to be invoked).

The network problem judgment decision logic and the network optimization decision logic in the embodiment of the application can be represented in the form of a decision tree, and the decision tree is a decision analysis method for judging the root of a problem or the feasibility of a problem solving scheme by constructing a series of related decision logics on the basis of the known occurrence probability of various conditions, and is a graphical method for intuitively applying probability analysis. This decision branch is called a decision tree because it is drawn to resemble a branch of a tree. In machine learning, a decision tree is a predictive model that represents a mapping between object attributes and object values.

The following describes a specific process of the network optimization method provided in the embodiment of the present application in detail with reference to fig. 4 to fig. 6.

Fig. 4 illustrates a method 400 for constructing a knowledge graph according to an embodiment of the present application. The method 400 may be applied to the system architecture 100 shown in fig. 1, and may also be applied to other system architectures, but the embodiment of the present application is not limited thereto. It should be appreciated that the method 400 may be performed by a data processing apparatus, as shown in FIG. 4, the method comprising:

s401, the data processing equipment determines a network information model.

The network information model may include at least one of: a network resource information model, a network environment information model, a user intention information model, or a network operation quality model, etc.

The input of the network resource information model can be information of any access network equipment, and the output can be the adjacent relation between the access network equipment and other surrounding access network equipment.

The input of the network environment information model can be information of any access network equipment, and the output can be density information of buildings around the access network equipment.

The input of the user intention information model is the subjective intention of the user, and the output is the objective intention of the user.

The input of the network operation quality model can be optimized network operation data, and the output can be that the network operation data meets the standard or does not meet the standard.

S402, the data processing equipment acquires information of various network information models and stores the information into the knowledge graph in a knowledge form.

Illustratively, the information in the network resource information model includes at least one of: the network-related terminals, network elements, links, and the relationship between them, for example, the relationship between network elements may be a neighboring relationship (adjacent or non-adjacent) between cells. The information in the environmental information model includes at least one of: geographical environment information, climate environment information, or living environment information of the terminal or the network element, and the like. The information in the user intent information model includes at least one of: subjective intention of the user, objective intention, psychological information of the user, or the like. The information in the network operational quality model includes performance data that characterizes the operational quality of the network.

The knowledge form is a form which can embody the incidence relation among various kinds of information. Illustratively, the above knowledge form may be a triplet (entity 1, relationship, entity 2) or (entity, attribute value) form, or an N-tuple form.

And S403, the data processing equipment acquires the network operation quality evaluation standard and stores the network operation quality evaluation standard in a knowledge graph in a knowledge form.

The data processing equipment obtains a network operation quality evaluation standard according to the historical operation data of the network, and stores the network operation quality evaluation standard in a knowledge graph in a triple or N-tuple mode. For example, the triplet form may be: (conditions of parameters or operating states, quality evaluation indexes, quality evaluation values).

S404, the data processing equipment determines a network problem judgment decision logic.

The data processing equipment acquires historical operation data of the network and network problems, and determines a problem judgment decision logic.

Illustratively, the network problem determination decision logic may be represented in the form of a decision tree, where the information of each node on the tree is: what network parameters or states are to be judged as a certain type of network problem, and the probability of judging as the type of problem.

S405, the data processing equipment stores the information of the network problem judgment logic into a knowledge graph in a knowledge form.

In conjunction with the above decision tree example, the data processing apparatus may store information for each node in the network problem determination decision tree in the form of triples or N-tuples in the knowledge-graph. For example, (a set of evaluation conditions for a parameter or state, a network problem classification probability, a network problem classification).

S406, the data processing equipment determines a network optimization decision logic.

The data processing device determines a network optimization decision logic based on historical network problems of the network and corresponding solutions.

Illustratively, the network optimization decision logic may be represented in the form of a decision tree, where the information for each node on the tree is: the network problem given at S405, the type of network optimization that needs to be performed, and the optimization algorithm that needs to be invoked.

S407, the data processing equipment stores the information of the network optimization decision logic in a knowledge graph in a knowledge form.

The information of the network optimization decision logic refers to a network optimization scheme corresponding to a network problem.

In conjunction with the above decision tree example, the data processing apparatus may store information in each node of the network optimization decision tree in the form of triples or N-tuples in the knowledge-graph. For example, (classification of network problems, current network parameters or operating conditions, categories of network optimizations that need to be performed, optimization algorithms that need to be invoked).

It will be appreciated that the above data processing apparatus, which may be a server, has a knowledge-graph generating tool deployed therein.

The knowledge graph provided by the embodiment of the application comprehensively considers various network information, realizes comprehensive collection and processing of the information, can reflect the actual network operation environment more truly, and is beneficial to judgment of network problems and formulation of a network optimization scheme.

Fig. 5 illustrates another network optimization method 500 provided in the embodiment of the present application. The method 500 may be applied to the system architecture 100 shown in fig. 1, and may also be applied to other system architectures, but the embodiment of the present application is not limited thereto. As shown in fig. 5, the method 500 includes:

and S501, the network management node collects data periodically or according to requirements.

The data comprises user complaint data, basic data of network resources, network operation data and the like. Wherein the user complaint data will be the source of the user intention analysis. The network base resource data will determine the scope of the search for the network problem. The network operational data will serve as a source of data for determining network problems. The following describes the network optimization method provided by the present application in detail by taking an example of a network management node obtaining user complaint data.

And S502, analyzing the objective intention of the user by the network management node according to the subjective complaint information of the user.

For the specific process, reference may be made to the above description for obtaining the objective intention of the user, which is not described herein again.

S503, the network management node determines the basic resource data of the target network based on the objective intention information of the user.

The basic resource data includes information of an access network device where the user is located, information of a cell, or location information of the cell.

S504, the network management node acquires information related to the target network from the knowledge graph based on the basic resource data.

Illustratively, the information obtained by the network management node from the knowledge-graph includes: the network performance of the target network element may be determined based on the relationship between the target network element and the surrounding network elements (e.g., the target network element is adjacent to or not adjacent to a plurality of surrounding network elements, etc.), network parameters for reflecting network characteristics (e.g., cell throughput or average interference level of the target network element, etc.), environmental characteristics around the network element (e.g., building density around the target network element, etc.), network operation data of the target network element, and so on.

And S505, the network management node autonomously judges the network problem based on the network problem judgment decision logic in the knowledge graph and determines the network problem.

The above network problem determination process may refer to the related description of S302 in the method 300, and is not described herein again.

And S506, the network management node judges whether the network problem needs to be optimized, if so, the step enters S507, and if not, the step enters S501.

S507, the network management node automatically generates a network optimization scheme based on the network optimization decision logic in the knowledge graph.

The network optimization scheme comprises network parameters needing to be adjusted and specific adjustment values.

S508, the network management node sends the optimization scheme to the network element in the target network so as to carry out network optimization execution.

The network management node may also proceed to S501 to collect various types of data continuously.

In the embodiment of the application, after the network management node acquires various data, the triggering of network optimization, the judgment of problems and the formulation and implementation of a network optimization scheme can be autonomously performed based on information in a knowledge graph, manual intervention is not needed, the defect that network optimization is carried out by relying on manual experience is effectively overcome, the labor cost and the optimization time cost are reduced, and meanwhile, the optimization efficiency is improved.

Fig. 6 illustrates a method 600 for updating a knowledge graph according to an embodiment of the present application. The method 600 may be applied to the system architecture 100 shown in fig. 1, and may also be applied to other system architectures, but the embodiment of the present application is not limited thereto. As shown in fig. 6, the method 600 includes:

and S601, the network management node evaluates the optimized network operation data obtained in the S408 according to the network operation quality evaluation standard in the knowledge graph.

For a specific evaluation process, reference may be made to the above description of the method 300 regarding the evaluation of the network operation quality, and details are not repeated here.

S602, the network management node judges whether the optimization effect meets the network operation quality evaluation standard. If the result is consistent with the preset value, the optimization is finished; otherwise, S603 is performed.

S603, the network management node selects other network optimization schemes to carry out network optimization, and updates 'network optimization decision logic' in the knowledge graph.

It should be understood that if there is no other network optimization scheme for the currently determined network problem, S603 is not performed, and S604 is performed directly.

S604, the network management node sequentially selects network optimization schemes corresponding to other network problems to perform network optimization, and updates 'network problem judgment decision logic' in the knowledge graph.

The updating process of S603 and S604 may specifically refer to the related description in the method 300, and is not described herein again.

And S605, finishing the optimization of the knowledge graph.

According to the method and the device, the knowledge in the knowledge graph is updated according to the evaluation result of the network operation quality, so that the knowledge graph is more time-efficient and dynamic, and the effect of network optimization is further improved.

The following describes in detail a specific process of another network optimization method provided in the embodiment of the present application with reference to fig. 7.

Fig. 7 illustrates another network optimization method 700 provided in the embodiment of the present application. The method 700 may be applied to the system architecture 100 shown in fig. 1, and may also be applied to other system architectures, but the embodiment of the present application is not limited thereto. As shown in fig. 7, the method 700 includes:

and S701, preparing a knowledge graph before optimization.

This step can be referred to the description of the method 400 above and will not be described here.

S702, network optimization process.

This step can be referred to the description of the method 500, and is not repeated here.

And S703, updating the optimized knowledge graph.

This step can be referred to the description of the method 600, and is not repeated here.

According to the method and the device, the network problems and the corresponding network optimization scheme are determined by obtaining the first network operation data of the target network and according to the first network operation data and the knowledge graph prepared before optimization. Therefore, the network element in the target network can complete network optimization based on the determined network optimization scheme, the problem that the network optimization scheme is determined by depending on manual experience is effectively solved, labor cost and optimization time cost are reduced, and meanwhile network optimization efficiency is improved.

The following describes in detail a network optimization method provided in the embodiments of the present application, by taking an optimization process of an inter-cell interference-type network problem as an example.

1. Data acquisition phase

(1) The network management node periodically collects network operation data or user complaint data. In the embodiment of the present application, for example, the user complaint data is taken as an example, and a plurality of pieces of user complaint data about the network speed, which are collected by the network management node, are shown in table one:

watch 1

Intention initiator	Subjective intention
		User A	"very poor net speed, how not yet resolved! "
User B	'what should be if the net is not frequently used'
		…	……

(2) And the network management node translates the collected subjective intention of the user into an objective intention. The expression form of objective intention is shown in table two:

watch two

Intention initiator	Objective intention
		User A	[ speed of network, difference]Solve the problem of bad angry]
User B	[ speed of network, difference]Solve the problem of poor network speed [ dissatisfaction]
		…	……

The content indicated in the second table [ ] "is the objective intention of the user, and the objective intention of the user is obtained by translating the complaint information of the user (i.e. the subjective intention of the user) by using a natural language processing technology through the network management node.

(3) The network management node obtains basic resource data (e.g., a base station or a cell) according to the user information as shown in table three: watch III

Intention initiator	Belonging base station (gNBId)	Belonging cell (NrCellId)
			User A	Ran001/gNB001	gNB001/NrCell001
User B	Ran001/gNB002	gNB002/NrCell001
			…	……	……

(4) And the network management node acquires data related to the target network from the network information in the knowledge graph based on the basic resource information.

Illustratively, the network configuration resource information in the network information in the knowledge-graph includes a relationship between network elements, for example, a neighboring relationship between cells, as shown in table four:

watch four

Cell 1 (Nr Cell Id)	Cell 2 (NrCellId)	Relationships between
			GNB001/NrCell001	GNB002/NrCell001	Adjacent to each other
GNB001/NrCell001	GNB002/NrCell002	Are not adjacent to each other
			……	……	……

Illustratively, the network configuration resource information in the network information in the knowledge-graph includes network performance parameters, e.g., important performance parameters such as cell throughput, as shown in table five:

watch five

Illustratively, the network environment information of the network information in the knowledge graph includes network element surrounding environment characteristics, such as the density of buildings surrounding the base station, and the like, as shown in table six:

watch six

Base station (gNB Id)	Density of surrounding buildings	……
			Ran001/gNB 001	High density (45%)	……
Ran001/gNB 002	High density (52%)	……
			……	……	……

2. Network problem judging and optimizing stage

(1) And determining the network problem based on the network problem judgment decision logic in the knowledge graph. For example, according to a pre-constructed network problem judgment decision logic, by analyzing various types of data obtained in the above steps, it can be judged that the network problem is inter-cell interference. The network problem judgment decision logic is stored in a knowledge graph in a form of triplets (user average signal to interference plus noise ratio, SINR) less than or equal to 0 decibel (dB), 0.7, interference-type problem), and the triplets respectively represent: network problem judgment conditions, network problem classification probabilities and judged network problems.

The information included in the network problem determination decision logic will be described in detail below with reference to fig. 8, which is an example of a network problem determination decision tree.

Fig. 8 illustrates a network problem determination decision tree provided in an embodiment of the present application. As shown in fig. 8, the network problem determination conditions are: the average SINR of the users is less than or equal to 0dB, and the network problems corresponding to the problem judgment conditions comprise: interference-like problems, capacity-like problems, no network problems, and other problems, wherein interference-like problems include: inter-cell interference problems, intra-cell interference problems, and other interference problems.

(2) And after the network problem is judged, determining a network optimization scheme for network optimization.

(3) Based on a network optimization decision logic in a knowledge graph, a power control optimization algorithm indicated by an inter-cell interference problem (the power of a main cell is lower) is preferentially selected, and a power control algorithm based on deep reinforcement learning (DQN) is preferentially invoked.

The information included in the network optimization decision logic will be described in detail below with reference to fig. 9, which is an example of a network optimization decision tree.

Fig. 9 illustrates a network optimization scheme decision tree provided in an embodiment of the present application. As shown in fig. 9, the optimization algorithm that can be invoked by the inter-cell interference-like problem includes: a power control algorithm, a load balancing algorithm, or other algorithms. Wherein the power control algorithm comprises: DQN based power control algorithm, power control algorithm B, power control algorithm a, or other kind of power control algorithm.

It should be understood that the above-mentioned callable optimization algorithms can be understood as multiple optimization schemes for solving the cell interference problem, and multiple optimization algorithms can be sorted according to the success rate for solving the cell interference problem, and the corresponding optimization algorithms are sequentially selected according to the sequence of the success rate from the highest to the lowest for network optimization.

Illustratively, the DQN-based power control algorithm is designed with statistical data of parameters such as SINR, path loss, throughput and transmit power of users in each cell as a state, with adjustment operation of transmit power of base stations in each cell as an action, with a goal of maximizing total system throughput on the premise of fair allocation of network resources, and is trained in advance and stored in a knowledge graph as a network optimization scheme.

(4) After the relevant network elements adopt corresponding algorithms to carry out network optimization, the transmitting power of each cell base station is adjusted to be proper.

3. Optimization effect evaluation and algorithm learning stage

(1) The effect evaluation index of the power control network optimization is edge user throughput, total throughput or total energy consumption, after algorithm optimization, the edge user throughput and the total throughput are improved, the total energy consumption is reduced, if the transmission power of each optimized cell base station reaches a threshold value condition, the optimization requirement is met, and the optimization task is completed.

(2) If the network optimization is effective but the effect is not obvious, the network optimization scheme is considered to be not appropriate. The network management node can adopt other network optimization schemes corresponding to the network problem to carry out network optimization until the optimization effect is obvious, namely the network optimization scheme meets the network operation quality evaluation standard, and the network optimization scheme is updated into a scheme with the highest probability of solving the network problem in the network optimization decision logic and stored in the knowledge graph.

(3) If the network optimization has no effect or is poor, the judgment of the network problem is considered to be wrong. The network management node can re-determine the network problem and perform network optimization by adopting a network optimization scheme corresponding to the network problem until the optimization effect is obvious, namely the network optimization scheme meets the network operation quality evaluation standard, and the network problem is updated into a scheme with the highest probability in the network problem logic and stored into the knowledge graph.

It should be understood that the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

The network optimization method according to the embodiment of the present application is described in detail above with reference to fig. 3 to 9, and the communication device according to the embodiment of the present application is described in detail below with reference to fig. 10 and 11.

Fig. 10 illustrates a communication apparatus 1000 according to an embodiment of the present application, where the apparatus 1000 includes: a transceiver module 1010 and a processing module 1020.

The transceiver module 1010 is configured to: acquiring first network operation data of a target network, wherein the first network operation data is used for reflecting the current operation state of the target network; the processing module 1020 is configured to: determining a first network problem based on the first network operation data, network problem judgment decision logic in a knowledge graph and network information in the knowledge graph, wherein the network problem judgment decision logic is used for representing a judgment condition of the network problem and the network problem corresponding to the judgment condition; and determining a first network optimization scenario for the first network problem based on network optimization decision logic in the knowledge graph, the network optimization decision logic being for representing an optimization scenario for a network problem; the transceiver module 1010 is further configured to: and sending the first network optimization scheme to a network element in the target network.

Optionally, the network problem judgment information includes a network problem judgment decision logic and network information, and the network problem judgment logic is configured to indicate a judgment condition of a network problem and a network problem corresponding to the judgment condition.

Optionally, the transceiver module 1010 is further configured to: acquiring second network operation data of the target network, wherein the second network operation data is operation data of the target network after optimization according to the first network optimization scheme; the processing module 1020 is further configured to: judging whether the second network operation data meets the network operation quality evaluation standard in the knowledge graph; and if the second network operation data does not accord with the network operation quality evaluation standard, determining whether a second network optimization scheme of the first network problem exists, wherein the network operation data optimized by the target network according to the second network optimization scheme accords with the network operation quality evaluation standard.

Optionally, the first network optimization scheme is a scheme with the highest power among a plurality of network optimization schemes for solving the first network problem; the processing module 1020 is further configured to: and sequentially performing network optimization according to the sequence of the success rate of the plurality of network optimization schemes for solving the first network problem from large to small until the second network optimization scheme exists.

Optionally, the processing module 1020 is further configured to: and if the second network optimization scheme does not exist, re-determining the network problem of the target network.

Optionally, the first network problem is a network problem with the highest accuracy among the plurality of network problems determined according to the network problem judgment decision logic; the processing module 1020 is further configured to: and sequentially determining the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists, wherein the network operation data of the target network after being optimized according to the target network optimization scheme meets the network operation quality evaluation standard.

Optionally, the processing module 1020 is further configured to: and under the condition that the second network operation data does not meet the network operation quality evaluation standard, updating the network optimization decision logic by taking the success rate of the second network optimization scheme as the highest-power scheme in a plurality of network optimization schemes for solving the first network problem, and/or updating the network problem judgment decision logic by taking the accuracy rate of the target network problem as the highest-accuracy network problem in the network problem judgment decision logic.

Optionally, the processing module 1020 is further configured to: natural language processing is carried out on complaint data of a user to obtain objective intention information of the user, wherein the complaint data of the user is used for reflecting the network state of the target network; and determining basic resource data of the target network based on the objective intention information of the user; the transceiver module 1010 is further configured to: and acquiring first network operation data of the target network from the network information in the knowledge graph based on the basic resource data.

It should be appreciated that the apparatus 1000 herein is embodied in the form of functional modules. The term module herein may refer to an Application Specific Integrated Circuit (ASIC), an electronic circuit, a processor (e.g., a shared, dedicated, or group processor) and memory that execute one or more software or firmware programs, a combinational logic circuit, and/or other suitable components that support the described functionality. In an optional example, it may be understood by those skilled in the art that the apparatus 1000 may be specifically a network management node in the foregoing embodiment, or functions of the network management node in the foregoing embodiment may be integrated in the apparatus 1000, and the apparatus 1000 may be configured to execute each process and/or step corresponding to the network management node in the foregoing method embodiment, and details are not described herein again to avoid repetition.

The device 1000 has a function of implementing corresponding steps executed by the network management node in the method; the above functions 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 functions described above. For example, the transceiver module 1010 may be a communication interface, such as a transceiver interface.

In the embodiment of the present application, the apparatus 1000 in fig. 10 may also be a chip or a chip system, for example: system on chip (SoC). Correspondingly, the transceiver module 1020 may be a transceiver circuit of the chip, and the application is not limited herein.

Fig. 11 illustrates a network optimization device 1100 according to an embodiment of the present application. The device 1100 includes a processor 1111, a transceiver 1120, and a memory 1130. The processor 1110, the transceiver 1120, and the memory 1130 are in communication with each other through an internal connection path, the memory 1130 is configured to store instructions, and the processor 1110 is configured to execute the instructions stored in the memory 1130 to control the transceiver 1120 to transmit and/or receive signals.

It should be understood that the apparatus 1100 may be embodied as a network management node in the foregoing embodiments, or functions of the network management node in the foregoing embodiments may be integrated in the apparatus 1100, and the apparatus 1100 may be configured to perform each step and/or flow corresponding to the network management node in the foregoing method embodiments. Alternatively, the memory 1130 may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information. The processor 1110 may be configured to execute the instructions stored in the memory, and when the processor executes the instructions, the processor may perform the steps and/or processes corresponding to the data processing apparatus in the above method embodiments.

It should be understood that, in the embodiments of the present application, the processor may be a Central Processing Unit (CPU), and the processor may also be other general processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

In implementation, the steps of the method 300 may be performed by integrated logic circuits of hardware or instructions in the form of software in a processor. The steps of a method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in a processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in a memory, and a processor executes instructions in the memory, in combination with hardware thereof, to perform the steps of the above-described method. To avoid repetition, it is not described in detail here.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for network optimization, comprising:

the method comprises the steps that a network management node obtains first network operation data of a target network, wherein the first network operation data are used for reflecting the current operation state of the target network;

the network management node determines a first network problem based on the first network operation data and network problem judgment information in the knowledge graph;

the network management node determines a first network optimization scenario for the first network problem based on network optimization decision logic in the knowledge-graph, the network optimization decision logic being used to represent an optimization scenario for a network problem;

and the network management node sends the information of the first network optimization scheme to a network element in the target network.

2. The method according to claim 1, wherein the network problem judgment information includes network problem judgment decision logic and network information, the network problem judgment logic is configured to indicate a judgment condition of a network problem and a network problem corresponding to the judgment condition, and the network information includes network configuration resource information or network environment information.

3. The method of claim 1 or 2, wherein after the network management node sends the first network optimization scheme to a network element in the target network, the method further comprises:

the network management node acquires second network operation data of the target network, wherein the second network operation data is operation data of the target network after optimization according to the first network optimization scheme;

the network management node judges whether the second network operation data meets the network operation quality evaluation standard;

if the second network operation data does not accord with the network operation quality evaluation standard, the network management node determines whether a second network optimization scheme of the first network problem exists, and the network operation data of the target network after being optimized according to the second network optimization scheme accords with the network operation quality evaluation standard.

4. The method of claim 3, wherein the first network optimization solution is a most powerful solution of a plurality of network optimization solutions to the first network problem.

5. The method of claim 4, wherein the determining whether the second network optimization scheme exists comprises:

and the network management node sequentially performs network optimization according to the sequence of the success rate of the plurality of network optimization schemes for solving the first network problem from high to low until the second network optimization scheme exists.

6. The method of claim 5, further comprising:

and if the second network optimization scheme does not exist, re-determining the network problem of the target network.

7. The method of claim 6, wherein the first network problem is a network problem with a highest accuracy among the plurality of network problems determined by the network problem determination decision logic.

8. The method of claim 7, wherein the re-determining the network problem with the target network comprises:

and the network management node sequentially determines the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists, wherein the network operation data of the target network after the target network is optimized according to the target network optimization scheme accords with the network operation quality evaluation standard.

9. The method of claim 8, further comprising:

and under the condition that the second network operation data does not meet the network operation quality evaluation standard, updating the network optimization decision logic by taking the second network optimization scheme as the most powerful scheme in a plurality of network optimization schemes for solving the first network problem.

10. A communications apparatus, comprising:

the system comprises a transceiving module, a first network operation module and a second network operation module, wherein the transceiving module is used for acquiring first network operation data of a target network, and the first network operation data is used for reflecting the current operation state of the target network;

the processing module is used for determining a first network problem based on the first network operation data and network problem judgment information in the knowledge graph, and the network problem judgment decision logic is used for representing a judgment condition of the network problem and the network problem corresponding to the judgment condition; and determining a first network optimization scenario for the first network problem based on network optimization decision logic in the knowledge graph, the network optimization decision logic being for representing an optimization scenario for a network problem;

the transceiver module is further configured to: and sending the information of the first network optimization scheme to a network element in the target network.

11. The communications apparatus according to claim 10, wherein the network problem determination information includes network problem determination decision logic and network information, the network problem determination logic is configured to indicate a determination condition of a network problem and a network problem corresponding to the determination condition, and the network information includes network configuration resource information or network environment information.

12. The communications device according to claim 10 or 11, wherein the transceiver module is further configured to:

acquiring second network operation data of the target network, wherein the second network operation data is operation data of the target network after optimization according to the first network optimization scheme;

the processing module is further configured to: judging whether the second network operation data meets the network operation quality evaluation standard or not; and if the second network operation data does not accord with the network operation quality evaluation standard, determining whether a second network optimization scheme of the first network problem exists, wherein the network operation data optimized by the target network according to the second network optimization scheme accords with the network operation quality evaluation standard.

13. The communications apparatus of claim 12, wherein the first network optimization scheme is a most powerful scheme of a plurality of network optimization schemes that solve the first network problem.

14. The communications apparatus of claim 13, wherein the processing module is further configured to:

and sequentially performing network optimization according to the sequence of the success rate of the plurality of network optimization schemes for solving the first network problem from large to small until the second network optimization scheme exists.

15. The communications apparatus of claim 14, wherein the processing module is further configured to:

and if the second network optimization scheme does not exist, re-determining the network problems existing in the target network.

16. The communications apparatus of claim 15, wherein the first network problem is a network problem with a highest accuracy among the plurality of network problems determined by the network problem determination decision logic.

17. The communications apparatus of claim 16, wherein the processing module is further configured to:

and sequentially determining the network optimization schemes of the network problems according to the sequence of the accuracy rates of the network problems from large to small until a target network optimization scheme of a target network problem exists, wherein the network operation data of the target network after being optimized according to the target network optimization scheme meets the network operation quality evaluation standard.

18. The communications apparatus of claim 17, wherein the processing module is further configured to: