KR20210056272A - Optimization for network data analytic device - Google Patents

Abstract

An optimization method for a network data analysis function device is disclosed. The optimization method allows the network data analysis function device to collect network data according to an analysis request of network data received from a consumer network function device and provide analysis information, which is an analysis result for the network data, to a consumer network function device. The network data analysis function device can receive a feedback about the analysis information from the consumer network function device and adjust the analysis method of the network data based on the feedback.

Description

Optimization method of network data analysis function device {OPTIMIZATION FOR NETWORK DATA ANALYTIC DEVICE}

The present invention relates to a method for optimizing a network data analysis function device.

Recently, while providing various services and low-latency networks, the complexity of network operation is increasing. For example, the network must support slices, vertical services, and flexible combinations of various services are required. As the complexity of such a network increases, there is a limit to the existing humans operating the network.

Thus, a separate network data analysis function device for analyzing network data generated in a network has been proposed. In a complex network, network data is generated in the same form as big data, but there is a need for a method to efficiently analyze network data and derive better quality analysis information.

The present invention provides a method of optimizing a network data analysis function device that analyzes network data based on feedback from a network function device that consumes analysis information of network data.

The present invention evaluates that the network data analysis function device performs analysis information on network data, and improves the performance of the network data analysis function device by changing an analysis method when analyzing network data based on the evaluation result. Provides a way.

According to an embodiment of the present invention, a method for optimizing a network data analysis function device includes: receiving a network data analysis request message from a consumer network function device that intends to use the analysis information of the network data; Collecting first network data for network analysis based on the analysis request message from a first provider network function device; Generating analysis information on network data by analyzing the collected network data based on a method of analyzing network data; Providing an analysis response message including analysis information of the network data to a consumer network function device; Receiving feedback on the analysis information of the network data from the consumer network function device; Collecting second network data from a second provider network function device to evaluate analysis information of the network data; And changing a method of analyzing network data performed by the device for analyzing network data based on at least one of the received feedback and second network data.

The generating of the analysis information may include selecting at least one analysis model from among a plurality of analysis models to generate analysis information of the network data.

In the changing step, the analysis model used to analyze the network data may be replaced, deleted, updated, additional training, or an analysis model added.

The changing may include changing at least one of a type of an analysis model used to analyze the network data, a weight of a layer, a number of layers, and an order of layers, or replacing a layer.

The changing may include filtering the first network data collected from the first network function device, and the filtering may include at least one of temporal filtering, spatial filtering, application filtering, network filtering, and slice filtering.

The changing may include changing, replacing, adding, or filtering the training data used when training the analysis model that generated the analysis information of the network data.

The changing may include increasing the number of times of learning of the analysis model, changing the type of the analysis model, or changing the properties of the analysis model while maintaining the type of the analysis model.

The feedback includes whether the consumer network function device has used the analysis information of the network data, information on the performance change of the consumer network function device when the analysis information of the network data is applied, and the analysis information of the network data. It may include at least one of a satisfaction level, a use case that is a use purpose of the analysis information of the network data, and temporal or spatial information about the use of the network data.

Receiving the analysis request message may include, when there are a plurality of network data analysis apparatuses, the consumer network function apparatus selects at least one network data analysis apparatus for generating analysis information of network data among network data analysis apparatuses, or Alternatively, a network function storage device may select, and each of the selected network data analysis devices may receive the analysis request message.

The step of changing the analysis method includes performing an evaluation on the analysis information of the first network data based on the feedback and the second network data, and based on the evaluation on the analysis information of the first network data, the You can change the analysis method of network data.

In the network data analysis function device according to an embodiment of the present invention, the network data analysis function device includes a processor, and the processor includes a network data analysis request message from a consumer network function device that intends to use the analysis information of the network data. Network data by receiving, and collecting first network data for network analysis based on the analysis request message from a first provider network function device, and analyzing the collected network data based on a method of analyzing network data. Generates analysis information for, and provides an analysis response message including analysis information of the network data to a consumer network function device, receives feedback on the analysis information of the network data from the consumer network function device, and receives the network data Network data performed by the network data analysis function device based on at least one of the received feedback and second network data, and collects second network data from the second provider network function device to evaluate the analysis information of the data You can change the method of analysis.

The processor may select at least one analysis model from among a plurality of analysis models to generate analysis information of the network data.

The processor may change an analysis method of network data by replacing, deleting, updating, additional training, or adding an analysis model used to analyze the network data.

The processor may change the analysis method of network data by changing at least one of a type of an analysis model used to analyze the network data, a weight of a layer, a number of layers, and an order of layers, or by replacing layers.

The processor may filter the first network data collected from the first network function device, and the filtering may include at least one of temporal filtering, spatial filtering, application filtering, network filtering, and slice filtering.

The processor may change the method of analyzing network data by changing, replacing, adding, or filtering training data used to train the analysis model that generated the analysis information of the network data.

The processor may increase the number of times of learning the analysis model, change the type of the analysis model, or change the properties of the analysis model while maintaining the type of the analysis model.

The feedback includes whether the consumer network function device has used the analysis information of the network data, information on the performance change of the consumer network function device when the analysis information of the network data is applied, and the analysis information of the network data. It may include at least one of a satisfaction level, a use case that is a use purpose of the analysis information of the network data, and temporal or spatial information about the use of the network data.

The processor, when the plurality of network data analysis devices, the consumer network function device selects at least one network data analysis device for generating network data analysis information from among the network data analysis devices, or the network function storage device After selecting, each of the selected network data analysis devices may receive the analysis request message.

The processor, based on the feedback and the second network data, performs an evaluation of the analysis information of the first network data, based on the evaluation of the analysis information of the first network data, the method of analyzing the network data Can be changed for.

According to an embodiment of the present invention, a network data analysis function device that analyzes network data may be optimized based on feedback from a network function device that consumes analysis information of network data.

According to an embodiment of the present invention, a network data analysis function is performed by changing the analysis method when the network data analysis function device evaluates the analysis information of the network data and analyzes the network data based on the evaluation result. The performance of the device can be improved.

1 is a diagram illustrating an operation of a device for analyzing network data according to an embodiment of the present invention.
2 is a flowchart illustrating a method of optimizing a device for analyzing network data according to an embodiment of the present invention.
3 is a diagram illustrating a processing procedure after receiving a feedback on analysis information of a network data analysis function device according to an embodiment of the present invention.
4 is a diagram illustrating a processing procedure of a device for analyzing network data according to an embodiment of the present invention.
5 is a diagram illustrating a process of processing when there are a plurality of network data analysis function devices according to an embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. The same reference numerals shown in each drawing indicate the same members.

Various changes may be made to the embodiments described below. The embodiments described below are not intended to be limited to the embodiments, and should be understood to include all changes, equivalents, and substitutes thereto.

Terms such as first or second may be used to describe various components, but these terms should be understood only for the purpose of distinguishing one component from other components. For example, a first component may be referred to as a second component, and similarly, a second component may be referred to as a first component.

The terms used in the examples are used only to describe specific embodiments, and are not intended to limit the embodiments. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate the presence of features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, but one or more other features. It is to be understood that the presence or addition of elements or numbers, steps, actions, components, parts, or combinations thereof does not preclude in advance.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiment belongs. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present application. Does not.

In addition, in the description with reference to the accompanying drawings, the same reference numerals are assigned to the same components regardless of the reference numerals, and redundant descriptions thereof will be omitted. In describing the embodiments, when it is determined that a detailed description of a related known technology may unnecessarily obscure the subject matter of the embodiment, a detailed description thereof will be omitted.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating an operation of a device for analyzing network data according to an embodiment of the present invention.

Referring to FIG. 1, a network data analysis function device (Network Data Analytics Function: NWDAF) 101 is provided in a consumer network function device 102 and at least one provider network function device. The corresponding first provider network function device 103-1, the second provider network function device 103-2, and a network storage function device 104 may be connected. In addition, although not shown in FIG. 1, the network data analysis function device 101 may be connected to an OAM function device (Operations, administration and management) 105.

The network data analysis function device 101 may receive an analysis request (or analysis subscription) for network data from the consumer network function device 102. The consumer network function device 102 refers to a network function device that intends to use analysis information on network data. Then, the network data analysis function device 101 may collect the first network data for analysis from the first provider network function device 103-1 based on the analysis request for the network data.

The network data analysis function device 101 may analyze the first network data collected from the first provider network function device 103-1. The first provider network function device 103-1 may mean a network function device that provides first network data necessary to generate analysis information of network data. The network data analysis function device 101 transmits the analysis information of the first network data collected from the first provider network function device 103-1 in response to the analysis request for the first network data to the consumer network function device 102. Can be provided to. In addition, the consumer network function device 102 may provide feedback on the analysis information of the first network data to the network data analysis function device 101.

The network data analysis function device 101 may collect second network data from the second provider network function device 103-2 in order to evaluate the analysis information of the first network data. The second provider network function device 103-2 may mean a network function device that provides second network data necessary for evaluating analysis information of the network data.

In this case, the first provider network function device 103-1 may be the same as or different from the second provider network function device 103-2. Also, the first network data may be the same as or different from the second network data.

The network data analysis function device 101 may evaluate analysis information of the first network data by using the second network data. In addition, the network data analysis function device 101 may change the analysis method for the first network data based on the evaluation of the analysis information of the first network data.

The operation of the network data analysis function device 101 will be described in more detail with reference to FIGS. 2 to 5.

2 is a flowchart illustrating a method of optimizing a device for analyzing network data according to an embodiment of the present invention.

In step (i), the consumer network function device 102 may transmit a network data analysis request to the network data analysis function device 101. At this time, the consumer network function device 102 may transmit a network data analysis request message (ex.Nnwdaf_AnalyticsInfo_Request) or a network data analysis subscription message (ex.Nnwdaf_AnalyticsSubscription_Subscribe) to the network data analysis function device 101.

The network data analysis request message or the network data analysis subscription message may include a type of a consumer network functional device (eg, AMF or SMF) and a use case indication. Alternatively, the network data analysis request message or the network data analysis subscription message may further include analysis filter information.

Here, the type is information for discriminating what kind of network function device the consumer network function device is. And, the use case indication is information representing a plurality of use cases. As an example, the use case indication identifies a purpose for analysis of network data requested by the consumer network function device 102.

In step (ii), the network data analysis function device 101 may collect the first network data from the provider network function device 103 or the OAM device 105. In addition, the network data analysis function device 101 analyzes the first network data collected according to the analysis request of the network data received from the consumer network function device 102 to generate analysis information on the first network data. I can.

In this case, the network data analysis function device 101 may identify an analysis model that generates analysis information for the first network data. For example, the network data analysis function device 101 may generate an analysis model by itself or may call an analysis model generated by another entity.

For example, the analysis model is a target of analytics reporting, analytics reporting parameters, analytics filter information, network function device type (NF Type), or use case indication. It can be determined based on at least one.

In this case, the target of the analysis reporting may mean a user terminal or a group of user terminals. In addition, the analysis reporting parameter may include a timestamp for the first network data collected by the network data analysis function device 101, a collection period for the first network data, and the like. The analysis filter information may include the location of the user terminal and slice information used by the user terminal. In addition, the type of the network function device may mean information for classifying the type of the consumer network function device 102. Also, the use case indication may mean information representing a plurality of use cases.

In step (iii), in order for the network data analysis function device 101 to evaluate a specific use case or an analysis model for the consumer network function device 102, it is necessary to obtain feedback from the consumer network function device 102. . Then, the network data analysis function device 101 may provide analysis information of the first network data to the consumer network function device 102 in response to a request for analysis of the network data received from the consumer network function device 102.

At this time, the response to the network data analysis request may include a response message (ex.Nnwdaf_AnalyticsInfo_Request_response) for the analysis request of network data or a notification message for the analysis subscription of network data (Nnwdaf_AnalyticsSubscription_Notify).

The response message for the analysis request of the network data or the notification message for the analysis subscription of the network data may include a feedback indication or an expiry time. The feedback indication is information for identifying whether it is necessary to receive feedback on the analysis information of the first network data from the consumer network function device 102. And the expiration time identifies the expiration time for obtaining feedback.

In step (iv), the consumer network function device 102 may receive analysis information on the first network data from the network data analysis function device 101. In addition, when the feedback indication is enabled in a response message to a request for analysis of network data including analysis information on the first network data or a notification message for a subscription to analysis of network data, the consumer network function device ( 102) may provide feedback on the analysis information of the first network data to the network analysis function device 101. In this case, the consumer network function device 102 may provide the feedback on the analysis information of the network data to the network data analysis device 101 through a feedback message (ex.Nnwdaf_AnalyticsInfo_Feedback or Nnwdaf_AnalyticsSubscription_Feedback).

Here, the feedback message is the analysis information ID (Analytics ID), whether the analysis information is used (Analytics Usage), the type of the consumer network function device (NF Type), a use case indication, and an evaluation value. , Notification Target Address (+ Notification Correlation Id), a timestamp for feedback, and the like.

The analysis information ID is information representing analysis information related to feedback. Whether the analysis information is used is information for identifying whether the consumer network function device 102 has used analysis information for the first network data. Whether to use the analysis information may be enabled when the consumer network function device 102 performs one or more actions based on the analysis information.

The type of the consumer network function device is information for identifying what kind of network function device (eg, AMF or SMF) the consumer network function device 102 is. The use case indication is information for identifying a purpose of an action performed by the consumer network function device 102 according to analysis information on the first network data. The use case indication represents a plurality of use cases.

According to an embodiment of the present invention, the use case indication described in step (i) and the use case indication described in step (iv) may be different. This is because the consumer network function device 102 does not perform an action on the use case indicated by the use case indication described in step (i).

The evaluation value represents the degree of satisfaction with the analysis information of the first network data obtained by the consumer network function device 102 from the network data analysis function device 101. The evaluation value is provided when the consumer network function device 102 evaluates analysis information of the first network data.

The notification target address and notification correlation information are information when a network data analysis is subscribed. The notification target address represents the address of the consumer network function device. The notification correlation ID is information for correlating notifications from the network data analysis function device 101 to the consumer network function device 102 when a subscription is applied. The timestamp for the feedback refers to the time point at which the consumer network function device generates the feedback.

In step (v), the network data analysis function device 101 may collect the second network data from the provider network function device 103 or the OAM device 105. The first network data may be the same as or different from the second network data. The provider network function device 103 described in step (ii) may correspond to the first provider network function device 103-1 in FIG. 1, and the provider network function device 103 described in step (v). ) May correspond to the second provider network function device 103-2 of FIG. 1.

In step (vi), the network data analysis function device 101 evaluates the analysis information of the first network data based on the feedback obtained from the consumer network function device 102 and the second network data collected in step (v). I can. In addition, the network data analysis function device 101 may change the analysis method used when generating analysis information of the first network data for a specific use case or a specific network function device based on the evaluation result. For example, the network data analysis function device 101 may update the analysis model used when generating analysis information of the first network data (eg, optimize or additionally train the analysis model).

3 is a diagram illustrating a processing procedure after receiving a feedback on analysis information of a network data analysis function device according to an embodiment of the present invention.

Referring to FIG. 3, the network data analysis function device 101 may receive a network data analysis request from the consumer network function device 102. Then, the network data analysis function device 101 may collect network data from the provider network function device 103 in response to the network data analysis request. In addition, the network data analysis function device 101 may select at least one of a plurality of analysis models (analysis model 1-analysis model N) and analyze the collected network data as the selected analysis model. The network data analysis function device 101 may generate analysis information of the network data by analyzing the network data using the selected analysis model.

According to an embodiment of the present invention, the network data analysis function device 101 may use a plurality of analysis models. And, each of the analysis models can be optimized for a specific use of the consumer network functional device 102. The network data analysis function device 101 is based on information (e.g., analysis reporting target, analysis filter, etc.) included in the analysis request of the network data transmitted by the consumer network function device 101 among a plurality of analysis models. Thus, you can select an appropriate analysis model.

Multiple analysis models may be essential to support different use cases for the same analysis information. As an example, requirements for an analysis model for the same analysis ID delivered by the consumer network function device 102 may be different for each use case. For example, in certain use cases, the processing delay that occurs when generating analysis information may be more important than the accuracy level, and in other use cases, the accuracy level may be relatively more important than the processing delay. .

So, according to an embodiment of the present invention, a plurality of analysis models may be supported. The network data analysis function device 101 may provide appropriate analysis information from an analysis model optimized for a use case for analysis information labeled with the same analysis ID.

According to an embodiment of the present invention, the network data analysis function device 101 includes information included in a network data analysis request message or a network data analysis subscription message transmitted from the consumer network function device 102 (for example, , Analysis reporting target, analysis reporting parameters, analysis filter information, network function device type, use case indication, etc.) can be categorized. The network data analysis function device 101 may recognize a use case of a network data analysis request according to a categorized result.

In addition, according to an embodiment of the present invention, a network using the network data collected from the second network function devices 103-2 or the OAM device 105 as well as the feedback received from the consumer network function device 102 An evaluation may be performed on the analysis information generated by the data analysis function device 101.

Further, according to an embodiment of the present invention, adjustment (eg, optimization or additional model training) may be performed on an analysis model that has generated analysis information based on the evaluation. Accordingly, the network data analysis function device 101 can provide more appropriate analysis information for a specific use in the consumer network function device 102.

As an example, analysis information of network data includes (i) slice load level information at the network slice instance level, (ii) observed service experience (ex.service experience for network slices, and Service experience for the application) statistical information or prediction information, (iii) load information of the network function device (ex. load statistics information or load prediction information about the network function device), (iv) network performance information (ex. network performance information) Statistical information or predicted information), (v) information related to the user terminal (e.g., user terminal mobility information (statistical information, predictive information), communication pattern, and user terminal communication information (statistical information, Prediction information), predicted behavior of the user terminal, abnormal behavior of the user terminal, etc.), (vi) congestion information of user data (statistical information, prediction information), (vii) persistence information of QoS (statistical information, prediction information) It can be identified by an analysis ID or the like indicating the like. For the aforementioned analysis information, the format and content of the network data to be collected by the network data analysis function device 101 may be different, and the content of the analysis information may be different.

Then, the network data analysis function device 101 may provide analysis information of the network data to the consumer network function device 102. In this case, the consumer network function device 102 may provide feedback on the analysis information of the network data to the network data analysis function device 101. According to an embodiment of the present invention, when the feedback indication is enabled, the consumer network function device 102 may provide feedback on the analysis information of the network data to the network data analysis function device 101. According to another embodiment of the present invention, regardless of the feedback indication, whenever the consumer network function device 102 receives the analysis information of the network data from the network data analysis function device 101, the feedback on the analysis information is transmitted to the network data. It can be provided to the analysis function device 101.

The network data analysis function device 101 may change the analysis method of network data based on the feedback provided by the consumer network function device 102. According to an embodiment of the present invention, the network data analysis function device 101 may analyze network data by selecting at least one analysis model from among a plurality of analysis models trained through machine learning. In this case, the analysis model may be a neural network, and various types of neural networks such as CNN, DNN, and SVM may be used. In this case, the analysis model may be trained by the network data function device 101 or may be trained by an entity other than the network data function device 101.

According to an embodiment of the present invention, changing the analysis method of network data may mean additionally training an analysis model used by the network data analysis function device 101 to analyze network data. In this case, as the evaluation value in the feedback of the network data decreases, the number of additional training of the analysis model may increase.

According to an embodiment of the present invention, changing the analysis method of network data is to change the analysis model used by the network data analysis function device 101 to analyze network data to another analysis model, or to analyze network data later. To do this, an analysis model may be added, or at least some or all of the selected plurality of analysis models may be replaced. Accordingly, an analysis model used when analyzing network data can be improved, and an analysis method can be updated.

DNN), 분석 모델의 종류는 유지하면서 분석 모델을 구성하는 레이어의 개수를 증가, 레이어를 교체하거나 또는 레이어별 특징값을 변경하는 것을 의미할 수 있다.According to an embodiment of the present invention, changing the analysis method of network data changes the type of analysis model used when analyzing the network data (ex. CNN)

DNN), while maintaining the type of the analysis model, it may mean increasing the number of layers constituting the analysis model, replacing the layers, or changing a feature value for each layer.

According to an embodiment of the present invention, changing the method of analyzing network data may mean changing, replacing, or adding a provider network function device 103 that provides network data. Alternatively, according to an embodiment of the present invention, changing the analysis method of network data does not use all of the network data collected from the provider network function device 103 as it is, but filtering (ex, temporal filtering-acquired at a specific time or Alternatively, it may mean using network data to which only network data generated at a specific time is used for analysis, or spatial filtering-acquired at a specific location or analyzed using only network data generated at a specific location) is applied. Alternatively, according to an embodiment of the present invention, changing the analysis method of network data includes the type, attribute (format, metadata, size, etc.), quantity, etc. of the network data collected from the provider network function device 103. It can mean changing.

According to an embodiment of the present invention, changing the analysis method of network data may mean changing/replacement of training data used when training an analysis model. Alternatively, according to another embodiment of the present invention, changing the analysis method of network data is to add or filter the training data used when training the analysis model (e.g., temporal filtering (learning generated by the user terminal at a specific time)) Data only), spatial filtering (only learning data generated in a specific place is used), application filtering (only learning data generated in a specific application used in the user terminal is used), network filtering (user terminal accesses a specific network or cell The training data generated when you do it, etc.), and slice filtering (learning data generated when a specific slice is assigned to the user terminal) can be performed.

According to an embodiment of the present invention, when the network data analysis function device 101 receives an analysis subscription rather than a network data analysis request from the consumer network function device 102, changing the analysis method of the network data is It may mean that the collection method for network data in the cycle is changed in the next cycle. Here, the method of collecting network data may include changing a temporal range, a spatial range, or a collection source.

Feedback on the analysis information may be provided to an entity training the analysis model. For example, when the network data analysis function device 101 trains an analysis model by itself, feedback on the analysis information provided by the consumer network function device 102 may be provided to the network data analysis function device 101. . However, in the case of training the analysis model in the network storage function device 104 connected to the network data analysis function device 101, the feedback on the analysis information provided by the consumer network function device 102 is the network storage function device ( 104).

As an example, the feedback is information on whether the consumer network function device 102 has used the analysis information of the network data, and information on the performance change of the consumer network function device when the analysis information of the network data is applied (e.g., processing speed , Whether or not the accuracy is changed or the degree of change), satisfaction with the analysis information of the network data, the purpose of using the analysis information of the network data, and temporal or spatial information about the use of the network data.

The network data analysis function device 101 may analyze network data to generate analysis information of network data. At this time, the network data analysis function device 101 does not self-evaluate the analysis information of the network data, but provides it to the consumer network function device 102 as it is, or evaluates itself, and only when the quality is higher than a certain level. It can be provided to the network function device 102. If the analysis information generated by the network data analysis function device 101 does not exhibit a certain level or higher quality, the network data analysis function device 101 is mentioned above before providing the analysis information to the consumer network function device 102. You can change the analysis method of one network data.

4 is a diagram illustrating a processing procedure of a device for analyzing network data according to an embodiment of the present invention.

Referring to FIG. 4, in step (1), the consumer network function device 102 may transmit a network data analysis request to the network data analysis function device 101. In step (2), the network data analysis function device 101 may collect the first network data from the first provider network function device 103-1. Here, the first network data may mean network data used for analysis of network data.

In step (3), the network data analysis function device 101 may analyze the first network data collected from the first provider network function device 103-1 to generate analysis information of the first network data. . Thereafter, the network data analysis function device 101 may provide analysis information of the first network data to the consumer network function device 102 that has transmitted the network data analysis request.

In step (4), the consumer network function device 102 may use the analysis information obtained from the network data analysis function device 101. In addition, the consumer network function device 102 may provide feedback on the analysis information to the network data function device 101.

In step (5), the network data analysis function device 101 may collect second network data from the second provider network function device 103-2. The second network data may mean network data for evaluating analysis information of the first network data.

In step (6), the network data analysis function device 101 may self-evaluate the analysis information of the first network data using the feedback and the second network data obtained from the consumer network function device 102. In addition, the network data analysis function device 101 may change the analysis method of the first network data based on the evaluation information self-evaluated for the analysis information of the first network data.

5 is a diagram illustrating a process of processing when there are a plurality of network data analysis function devices according to an embodiment of the present invention.

Referring to FIG. 5, a case in which there are N network data analysis function devices will be described. The consumer network function device 102 may select any one of the N network data analysis function devices 101-1 to 101-N to transmit a network data analysis request in order to use the analysis information of the network data. . At this time, the consumer network function device 102 obtains a list of N network data analysis function devices 101-1 to 101-N from the network storage function device 104, and is suitable for requesting analysis of network data. At least one network data analysis function device may be selected. Alternatively, the consumer network function device 102 may contact the network storage function device 104 for a network data analysis device suitable for a network data analysis request to confirm. The network storage function device 104 may store addresses and profiles for the network function device. Here, the profile of the network function device may mean the type and type of the network function device.

Alternatively, the consumer network function device 102 may request network data analysis from different network data analysis devices for each analysis of the network data. For example, when requesting the analysis information A of network data, the consumer network function device 102 may request the network data analysis function device 101-1 for analysis information of the network data. In addition, when requesting the analysis information B of the network data, the consumer network function device 102 may request the network data analysis function device 101-2 for analysis information of the network data.

As an example, the consumer network function device 102 may request the same analysis information from a plurality of network data function analysis devices in order to request analysis information of the same network data. In addition, the consumer network function device 102 may use the analysis information after acquiring all analysis information from the plurality of network data function analysis devices. Alternatively, the consumer network function device 102 obtains all analysis information from a plurality of network data function analysis devices, and then provides at least one analysis information according to a specific criterion (accuracy, preference, time, space, application). You can choose and use it.

Meanwhile, the method according to the present invention is written as a program that can be executed on a computer and can be implemented in various recording media, such as a magnetic storage medium, an optical reading medium, and a digital storage medium.

Implementations of the various techniques described herein may be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or combinations thereof. Implementations may include a data processing functional device, e.g., a programmable processor, a computer, or a computer program product, i.e. an information carrier, e.g., machine-readable It may be implemented as a storage function device (computer readable medium) or a computer program tangibly embodied in a radio signal. Computer programs such as the above-described computer program(s) may be recorded in any type of programming language, including compiled or interpreted languages, and as a standalone program or in a module, component, subroutine, or computing environment. It can be deployed in any form, including as other units suitable for the use of. A computer program can be deployed to be processed on one computer or multiple computers at one site or to be distributed across multiple sites and interconnected by a communication network.

Processors suitable for processing a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. In general, the processor will receive instructions and data from read-only memory or random access memory or both. The elements of the computer may include at least one processor that executes instructions and one or more memory functional devices that store instructions and data. In general, a computer may include one or more mass storage functional devices that store data, such as magnetic, magnetic-optical disks, or optical disks, or receive data from or transmit data to them, or It can also be combined so as to be both. Information carriers suitable for embodying computer program instructions and data are, for example, semiconductor memory functional devices, for example magnetic media such as hard disks, floppy disks and magnetic tapes, Compact Disk Read Only (CD-ROM). Memory), optical media such as DVD (Digital Video Disk), magnetic-optical media such as floptical disk, ROM (Read Only Memory), RAM ( RAM, Random Access Memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), and the like. The processor and memory may be supplemented by or included in a special purpose logic circuit structure.

Further, the computer-readable medium may be any available medium that can be accessed by a computer, and may include both a computer storage medium and a transmission medium.

While this specification includes details of a number of specific implementations, these should not be construed as limiting to the scope of any invention or claimable, but rather as a description of features that may be peculiar to a particular embodiment of a particular invention. It must be understood. Certain features described herein in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination. Furthermore, although features operate in a particular combination and may be initially described as so claimed, one or more features from a claimed combination may in some cases be excluded from the combination, and the claimed combination may be a sub-combination. Or sub-combination variations.

Likewise, although operations are depicted in the drawings in a specific order, it should not be understood that such operations must be performed in that particular order or sequential order shown or that all illustrated operations must be performed in order to obtain a desired result. In certain cases, multitasking and parallel processing can be advantageous. In addition, the separation of the various functional device components of the above-described embodiments should not be understood as requiring such separation in all embodiments, and the program components and functional devices described are generally integrated together into a single software product or in multiple software products. It should be understood that it can be packaged.

On the other hand, the embodiments of the present invention disclosed in the specification and drawings are merely presented specific examples to aid understanding, and are not intended to limit the scope of the present invention. It is apparent to those of ordinary skill in the art that other modified examples based on the technical idea of the present invention may be implemented in addition to the embodiments disclosed herein.

Claims (20)

In the network data analysis function device optimization method,
Receiving a network data analysis request message from a consumer network function device that intends to use the network data analysis information;
Collecting first network data for network analysis based on the analysis request message from a first provider network function device;
Generating analysis information on network data by analyzing the collected network data based on a method of analyzing network data;
Providing an analysis response message including analysis information of the network data to a consumer network function device;
Receiving a feedback on the analysis information of the network data from the consumer network function device;
Collecting second network data from a second provider network function device to evaluate analysis information of the network data;
Changing a network data analysis method performed by the network data analysis function device based on at least one of the received feedback and second network data
Optimization method of a network data analysis function device comprising a. The method of claim 1,
Generating the analysis information,
An optimization method of a network data analysis function device for selecting at least one analysis model from among a plurality of analysis models to generate analysis information of the network data. The method of claim 1,
The changing step,
An optimization method of a network data analysis function device for replacing, deleting, updating, additional training, or adding an analysis model to an analysis model used to analyze the network data. The method of claim 1,
The changing step,
A method of optimizing a network data analysis function device for changing at least one of a type of an analysis model used to analyze the network data, a weight of a layer, a number of layers, and an order of layers, or replacing layers. The method of claim 1,
The changing step,
Filtering the first network data collected from the first network function device,
The filtering may include at least one of temporal filtering, spatial filtering, application filtering, network filtering, and slice filtering. The method of claim 1,
The changing step,
An optimization method of a network data analysis function device for changing, replacing, adding, or filtering training data used when training an analysis model that generates analysis information of the network data. The method of claim 1,
The changing step,
An optimization method of a network data analysis function device in which the number of times of learning of the analysis model is increased, the type of the analysis model is changed, or properties of the analysis model are changed while maintaining the type of the analysis model. The method of claim 1,
The feedback is,
Whether the consumer network function device uses the analysis information of the network data, information on the performance change of the consumer network function device when the analysis information of the network data is applied, satisfaction with the analysis information of the network data, the network A method for optimizing a network data analysis function device comprising at least one of a use case for use of data analysis information and temporal or spatial information for use of the network data. The method of claim 1,
Receiving the analysis request message,
When there are a plurality of network data analysis devices, the consumer network function device selects at least one network data analysis device that generates analysis information of network data from among the network data analysis devices, or a network function storage device selects the An optimization method of a network data analysis function device in which each of the selected network data analysis devices receives the analysis request message. The method of claim 1,
The step of changing the analysis method,
Performing evaluation on the analysis information of the first network data based on the feedback and the second network data,
An optimization method of a network data analysis function device for changing the analysis method of the network data based on an evaluation of the analysis information of the first network data. In the network data analysis function device,
The network data analysis function device includes a processor,
The processor,
Receiving a network data analysis request message from a consumer network function device that wants to use the network data analysis information,
Collecting first network data for network analysis based on the analysis request message from a first provider network function device,
Generate analysis information for network data by analyzing the collected network data based on a method of analyzing network data,
Providing an analysis response message including analysis information of the network data to a consumer network function device,
Receiving a feedback on the analysis information of the network data from the consumer network function device,
Collecting second network data from a second provider network function device to evaluate the analysis information of the network data,
A network data analysis function device for changing a network data analysis method performed by the network data analysis function device based on at least one of the received feedback and second network data. The method of claim 11,
The processor,
An optimization method of a network data analysis function device for selecting at least one analysis model from among a plurality of analysis models to generate analysis information of the network data. The method of claim 11,
The processor,
A network data analysis function device for changing an analysis method of network data by replacing, deleting, updating, additional training, or adding an analysis model to the analysis model used to analyze the network data. The method of claim 11,
The processor,
A network data analysis function device that changes an analysis method of network data by changing at least one of a type of an analysis model used to analyze the network data, a weight of a layer, a number of layers, and an order of layers, or by replacing layers. The method of claim 11,
The processor,
Filtering the first network data collected from the first network function device,
The filtering may include at least one of temporal filtering, spatial filtering, application filtering, network filtering, and slice filtering. The method of claim 11,
The processor,
A network data analysis function device for changing an analysis method of network data by changing, replacing, adding, or filtering training data used when training an analysis model that generated analysis information of the network data. The method of claim 11,
The processor,
A network data analysis function device for increasing the number of times of learning of the analysis model, changing the type of the analysis model, or changing properties of the analysis model while maintaining the type of the analysis model. The method of claim 11,
The feedback is,
Whether the consumer network function device uses the analysis information of the network data, information on the performance change of the consumer network function device when the analysis information of the network data is applied, satisfaction with the analysis information of the network data, the network A network data analysis function device comprising at least one of a use case for use of data analysis information and temporal or spatial information for use of the network data. The method of claim 11,
The processor,
When there are a plurality of network data analysis devices, the consumer network function device selects at least one network data analysis device that generates analysis information of network data from among the network data analysis devices, or a network function storage device selects the Each of the selected network data analysis devices receives the analysis request message. The method of claim 11,
The processor,
Performing evaluation on the analysis information of the first network data based on the feedback and the second network data,
A network data analysis function device that changes the analysis method of the network data based on an evaluation of the analysis information of the first network data.

Images