CN109787846A - A kind of 5G network service quality exception monitoring and prediction technique and system - Google Patents
A kind of 5G network service quality exception monitoring and prediction technique and system Download PDFInfo
- Publication number
- CN109787846A CN109787846A CN201910235520.3A CN201910235520A CN109787846A CN 109787846 A CN109787846 A CN 109787846A CN 201910235520 A CN201910235520 A CN 201910235520A CN 109787846 A CN109787846 A CN 109787846A
- Authority
- CN
- China
- Prior art keywords
- data
- qos
- service quality
- network service
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention proposes a kind of 5G network service quality exception monitoring and prediction technique and system, belongs to technical field of communication network.The system comprises: data acquisition module: for acquiring 5G network service quality data and network KPI performance monitoring data；Data processing module: for the network service quality data to be pre-processed and are marked；Qos data memory module: for storing the network service quality data after the label；Model training module: establishing Supervised machine learning model, and training obtains QoS exception monitoring device and QoS predicting abnormality device；QoS exception monitoring device: current 5G network service quality data are monitored；QoS predicting abnormality device: prediction future 5G network service quality data exception；Qos policy decision-making module: for marking and storing abnormal data, abnormal results are reported.The present invention can be the quality of service guarantee of the 5G network user, improve service quality.
Description
Technical field
The invention belongs to technical field of communication network, and in particular to a kind of 5G network service quality based on decision tree is abnormal
Monitoring and prediction technique and system.
Background technique
Network service quality (QoS, Quality of Service) is the necessary support for guaranteeing network performance, traditional net
Network quality of service guarantee is using Differentiated Services Diffserv model or integrated service Interserv model.Diffserv can not
Guarantee global optimization, Interserv model is related to complicated signaling control, therefore cause existing network can only provide as possible and
For service quality and the guarantee of service quality can not be provided.
It is connected in bulk device, under ultra high flux density, superelevation connection number density and ultrahigh-mobility scene, 5G network is such as
What meets user service there are huge challenge, and traditional network service quality framework can not adapt to complicated, dynamic 5G network and answer
Use scene.
Summary of the invention
In consideration of it, the present invention proposes a kind of 5G network service quality exception monitoring and prediction technique and system, for monitoring
Network service is abnormal, improves network service quality.
First aspect present invention proposes a kind of 5G network service quality exception monitoring and prediction technique, which comprises
S1, acquisition 5G network service quality data and network KPI performance monitoring data, the network service quality data packet
Include user terminal data, access network data, core network data；
S2, the network service quality data are pre-processed and is marked；
S3, the network service quality data after the label are stored to QoS data library；
S4, using described in the QoS data library mark after network service quality data as data set, be built with supervision
Machine learning model is trained the Supervised machine learning model using the data set, obtains QoS exception monitoring device
With QoS predicting abnormality device；
S5, using the current 5G network service quality data of QoS exception monitoring device real-time monitoring, the exception that will be monitored
Data are sent to qos policy decision-making module；
S6, future 5G network service quality data exception, the exception that will be predicted are predicted using the QoS predicting abnormality device
Data are sent to qos policy decision-making module；
S7, the qos policy decision-making module mark and store the abnormal data, update QoS data library, report abnormal knot
Fruit simultaneously makes decisions decision according to the abnormal data, and the decision is driven to determine to execute.
Optionally, in step S1: the user terminal data includes: the hardware data of user terminal, software model version,
The application of installation, terminal location, moving direction, speed, CPU, memory source, the alarm log of consumption；The access network data
It include: base station distribution, antenna channel mode, frequency spectrum use, physical resource virtual resource service condition, space interface signaling, alarm day
Will；The core network data includes: QoS of customer agreement, and network is sliced resource and uses, core network signalling, alarm log；
The network KPI performance monitoring data includes: network bandwidth, time delay, shake.
Optionally, the detailed process of step S2 is stated are as follows: to collecting clean, uniting for the network service quality data
One format, is marked the network service quality data in conjunction with the network KPI performance monitoring data, and the label includes
It is normal and abnormal.
Optionally, in the step S4, the Supervised machine learning model uses decision Tree algorithms.
Optionally, it states in step S6, the QoS predicting abnormality device is according to current 5G network service quality data and history 5G
Network service quality data are to predict the following 5G network service quality data exception, the history 5G network service quality data
The web-based history quality of service data record saved in the QoS data library.
Optionally, it states in step S7, the qos policy decision-making module marks and store the abnormal data specifically: from
It is dynamic that the result of the QoS exception monitoring device and the QoS predicting abnormality device is marked, and new flag data is stored in institute
QoS data library is stated, network service quality data in the QoS data library are updated.
Second aspect of the present invention provides a kind of 5G network service quality exception monitoring and forecasting system, the system comprises:
Data acquisition module: for acquiring 5G network service quality data and network KPI performance monitoring data, the network
Quality of service data includes user terminal QoS data, Access Network QoS data, core net QoS data；
Data processing module: for the network service quality data to be pre-processed and are marked；
Qos data memory module: for storing the network service quality data after the label；
Model training module: for using the network service quality data after being marked described in the QoS data library as having
The data set of supervision machine learning model is built with supervision machine learning model, carries out to the Supervised machine learning model
Training, obtains QoS exception monitoring device and QoS predicting abnormality device
QoS exception monitoring device: being used for the current 5G network service quality data of real-time monitoring, and the abnormal data monitored is sent out
It send to qos policy decision-making module；
QoS predicting abnormality device: for according in current 5G network service quality data and the Qos data memory module
History 5G network service quality data predict future 5G network service quality data exception, and the abnormal data predicted is sent to
Qos policy decision-making module；
Qos policy decision-making module: for marking and storing the abnormal data, abnormal results are reported and according to the exception
Data make decisions decision, and the decision is driven to determine to execute.
Optionally, the data acquisition module specifically includes:
User terminal data acquisition unit: for obtaining hardware data, the software model version of user terminal, installation is answered
With, terminal location, moving direction, speed, CPU, memory source, the alarm log of consumption；
Access network data acquisition unit: for obtaining, base station distribution data, antenna channel mode, frequency spectrum are used, physics provides
Source virtual resource service condition, space interface signaling, alarm log data；
Core network data acquisition unit: for obtaining QoS of customer agreement, network is sliced resource and uses, core net letter
It enables, alarm log data；
Network KPI performance monitoring data acquisition unit: for obtaining network bandwidth, time delay, shake data.
Optionally, the model training module constructs the Supervised machine learning model, training using decision Tree algorithms
The result is that a tree being made of node and branch, each non-leaf nodes indicate one in the data set
Attribute, some value of branch, that is, attribute or value section, a classification in each leaf node, that is, data set, table
Show that network service quality is abnormal or normal.
Optionally, the qos policy decision-making module is by the different of the QoS exception monitoring device and the QoS predicting abnormality device
Normal result is marked, and new flag data is stored in the QoS data library, updates network service in the QoS data library
Qualitative data.
The present invention proposed by the present invention proposes a kind of 5G network service quality exception monitoring and prediction technique, by collecting,
Storage marks, the QoS service qualitative data of the network terminal of analysis magnanimity, wireless access network, core net:
1) incidence relation of restructural web-based history event and network service quality；
2) it can be further formed network QoS management strategy, by soft to the real-time monitoring of current network service quality exception
Part, virtualization, Slice interface for network programming, realize the Automatic dispatching of Internet resources, to be the service of the 5G network user
Quality assurance is improved service quality；
3) following possible network service quality can be predicted extremely, is mentioned for the network planning and service quality optimization
For foundation.
Detailed description of the invention
It, below will be to needed in the technology of the present invention description in order to illustrate more clearly of technical solution of the present invention
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without any creative labor, it can also be obtained according to these attached drawings others
Attached drawing.
Fig. 1 is network service quality exception monitoring provided by the invention and prediction technique flow diagram；
Fig. 2 is user terminal data provided by the invention, the tag format for accessing network data, core network data；
Fig. 3 is network service quality exception monitoring provided by the invention and forecast and decision tree schematic diagram；
Fig. 4 is network service quality exception monitoring provided by the invention and forecasting system structural schematic diagram.
Specific embodiment
The present invention proposes a kind of 5G network service quality exception monitoring based on decision tree and prediction technique and system, passes through
Collect, label, storage, the analysis network terminal of magnanimity, wireless access network, core net QoS service qualitative data, using there is prison
Superintend and direct real-time monitoring and prediction that machine learning model realizes network service quality exception.
From the point of view of machine learning model, machine learning is divided into supervised learning, unsupervised learning and semi-supervised learning.
Under learning method with supervision, input data is referred to as " training data ", and every group of training data has a specific mark
Know or as a result, learning method with supervision establishes a study such as to " spam " " non-spam email " in Anti-Spam
Prediction result is compared by process with the actual result of " training data ", continuous to adjust prediction model meal place, until model
Prediction result reach an expected accuracy rate, supervised learning algorithm includes linear regression, decision tree, support vector machines
Deng.
In the study of unsupervised formula, data are not particularly identified, and learning model is to be inferred to some interior of data
In structure.Common application scenarios include study and cluster of correlation rule etc., and unsupervised learning algorithm includes K-means,
Hierarchical clustering etc..
Under semi-supervised learning mode, input data part is identified, and partially without being identified, this learning model can be with
For being predicted, but model is predicted firstly the need of the immanent structure of learning data reasonably to organize organization data.
Machine learning depends primarily on algorithm, computing capability and data, obtain after data can using online mode or
Offline mode is trained and calculates, and offline mode real-time is not high, it is contemplated that the real-time of network service quality guarantee is wanted
It asks, the present invention uses on-line study and training method.
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that disclosed below
Embodiment be only a part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
Those of ordinary skill's all other embodiment obtained without making creative work, belongs to protection of the present invention
Range.
Referring to Figure 1, the present invention proposes a kind of 5G network service quality exception monitoring and prediction technique, the method packet
It includes:
S1, acquisition 5G network service quality data and network KPI performance monitoring data, the network service quality data packet
Include user terminal data, access network data, core network data；
The user terminal data includes: the hardware data of user terminal, software model version, the application of installation, terminal
Position, moving direction, speed, CPU, memory source, the alarm log of consumption；
The access network data includes: base station distribution, antenna channel mode, frequency spectrum uses, physical resource virtual resource makes
With situation, space interface signaling, alarm log；
The core network data includes: QoS of customer agreement, and network is sliced resource and uses, core network signalling, alarm
Log；
Network KPI (the Key Performance Indication) performance monitoring data include: network bandwidth, when
Prolong, shake.
S2, the network service quality data are pre-processed and is marked；
To collect the network service quality data clean, unified format, supervised in conjunction with the network KPI performance
The network service quality data are marked in measured data, and the label includes normal and abnormal.
Specifically, pretreatment is to handle invalid value and missing values, and will to remove noise and extraneous data and correct mistake
Collected data unified standard, format converting formatting for data, being easy to subsequent processing.By network bandwidth,
The networks KPI performance monitoring data such as time delay, shake, respectively by pretreated user terminal data, access network data, core net
Data markers are at normal or abnormal.The certain KPI performance datas of network can monitor, such as rate, and time delay etc. can be direct
The user data of corresponding acquisition is marked according to these KPI datas of monitoring, accesses network data and core network data.
Fig. 2 is referred to, the tag format that user terminal data is listed in Fig. 2, accesses network data, core network data.Fig. 2
In, Feature1, Feature2 etc. are the attribute or feature of data set, corresponding to be classified as every attribute value or value section, mark
The class label of note has two classes, is Normal (normal) and Anomaly (exception).Network service after label can be passed through after label
Qualitative data learns marking convention, training prediction model, last automatic monitoring current network service quality and according to learning
Rule classification is to prediction result.
S3, the network service quality data after the label are stored to QoS data library；
Specifically, can be used relevant database or NoSQL database or file system stored,
S4, using described in the QoS data library mark after network service quality data as data set, be built with supervision
Machine learning model is trained the Supervised machine learning model using the data set, obtains QoS exception monitoring device
With QoS predicting abnormality device；
In the step S4, the Supervised machine learning model uses decision Tree algorithms.Adoptable algorithm further includes
Neural network or algorithm of support vector machine etc..
Decision Tree algorithms are a kind of supervised learning methods for classification, and decision Tree algorithms can be by the feature of training set data
It is inferred to decision rule, the result of input data is predicted by creation tree.It will be marked described in the QoS data library
Data set is generally divided into training dataset and test data set as data set by network service quality data afterwards, passes through instruction
Practice data set to generate decision tree, how to predict to export from input data study, test sample collection is reused to determine generation
Plan tree tests, corrects and repair down, generates in the data check Decision Tree Construction concentrated by test data preliminary
Those branches for influencing accuracy are wiped out, more accurate tree are generated, finally with the classification results of leaf node by rule
The monitoring result of (normal or abnormal) as new input network service quality data.
By taking C4.5 decision Tree algorithms as an example, illustrate the detailed process of step S4:
1) it calculates training data and concentrates the information gain-ratio of each attribute, and construct decision-tree model.
Specifically, sorting algorithm is based on comentropy, comentropy is bigger, and the new information of representative information institute band is more, then two
There is significantly difference in data, thus belong to inhomogeneity；Comentropy is smaller, and the new information of representative information institute band is fewer, then and two
Data probably belong to one kind.If the value of attribute be it is continuous, first by Discretization for Continuous Attribute, after discretization, it is assumed that belong to
Property A attribute value have m it is discrete after value interval, then training dataset S is divided into C by the attribute value of attribute A1, C2...,
CmTotal m Sub Data Set, | Cp| indicate sample size in p-th of Sub Data Set, | S | sample is total in data set before indicating to divide
Quantity,
Remember P (Cp) it is classification subset CpThe frequency occurred in sample set S, P (Cp)=| Cp|/| S |, p=1,2 ..., m,
The entropy of sample set before division:
For wherein any attribute Ai, it is assumed that there is t different value aq, q=1,2 ..., t, according to AiDifferent values, can
S is divided into S1, S2..., StTotal t subset, while can be by C1, C2..., CmIt is divided into m*t subset, each subset Cpq
It indicates in Ai=aqUnder conditions of belong to the sample set of pth class, cross attribute AiThe entropy of sample set after division:
Wherein,Entropy is smaller, and the purity of subset division is higher.Pass through attribute Ai
The information gain of sample set after division are as follows:
InfoGain(S,Ai)=H (S)-H (S, Ai)
Information gain InfoGain (S, Ai) indicate uncertain decline degree after dividing.
Attribute AiDivision information content:
Continue division and create new node, passes through attribute AiThe information gain-ratio of sample set S after division are as follows:
The attribute of C4.5 decision Tree algorithms selection maximum information ratio of profit increase establishes initial decision tree from top to bottom, utilizes test
Data set carries out beta pruning to initial decision tree, and hooks of going out improve classification accuracy, obtain final decision tree-model.
Fig. 3, exception monitoring and forecast and decision tree schematic diagram are referred to, decision tree is the class being made of node and branch
The tree of flow chart, interior joint are divided into leaf node and non-leaf nodes again.The top layer of tree is first non-leaf section
Point --- root node is the initial position of decision tree.Each non-leaf nodes indicates some attribute in data set, branch
Some value of the i.e. described attribute or value section.A classification in each leaf node, that is, data set, i.e. expression network Service Quality
Amount is abnormal or normal.
2) the decision tree module built can be used as QoS exception monitoring device, and the input of current 5G network service quality is determined
Plan tree-model can monitor QoS exception；According to the history 5G in current 5G network service quality data and Qos data memory module
Network service quality data, which can be trained further, obtains QoS predicting abnormality device, predicts future 5G network service quality data exception.
S5, using the current 5G network service quality data of QoS exception monitoring device real-time monitoring, the exception that will be monitored
Data are sent to qos policy decision-making module；
S6, future 5G network service quality data exception, the exception that will be predicted are predicted using the QoS predicting abnormality device
Data are sent to qos policy decision-making module；
In the step S6, the QoS predicting abnormality device is according to current 5G network service quality data and history 5G network
Quality of service data predicts the following 5G network service quality data exception, and the history 5G network service quality data are described
The web-based history quality of service data record saved in QoS data library.
S7, the qos policy decision-making module mark and store the abnormal data, update QoS data library, report abnormal knot
Fruit simultaneously makes decisions decision according to the abnormal data, and the decision is driven to determine to execute.
In the step S7, the qos policy decision-making module marks and stores the abnormal data specifically:
Automatically the result of the QoS exception monitoring device and the QoS predicting abnormality device is marked, and by new label
Data are stored in the QoS data library, save as history 5G network service quality data, update network in the QoS data library
Quality of service data.The incidence relation of web-based history event and network service quality is established with this, and is trained and pre- in next step
It surveys and training foundation is provided.After the update of QoS data library when there is new network service quality data input, then by updated QoS
Database simultaneously predicts the exception in input data as new data set, training.
Qos policy decision-making module makes a policy also according to the abnormal results for monitoring or predicting, for example bandwidth not enough then increases
Add bandwidth, time delay is too long, accelerates queue queuing processing etc., and decision is driven to execute.
Fig. 4 is referred to, the present invention also provides a kind of 5G network service quality exception monitoring and forecasting system, the system packet
It includes:
Data acquisition module 410: described for acquiring 5G network service quality data and network KPI performance monitoring data
Network service quality data include user terminal QoS data, Access Network QoS data, core net QoS data；
Data processing module 420: for the network service quality data to be pre-processed and are marked；
Qos data memory module 430: for storing the network service quality data after the label；
Model training module 440: for using described in the QoS data library mark after network service quality data as
The data set of Supervised machine learning model is built with supervision machine learning model, to the Supervised machine learning model into
Row training, obtains QoS exception monitoring device and QoS predicting abnormality device
QoS exception monitoring device 450: the current 5G network service quality data of real-time monitoring, the abnormal number that will be monitored are used for
According to being sent to qos policy decision-making module；
QoS predicting abnormality device 460: for according to current 5G network service quality data and the Qos data memory module
In history 5G network service quality data predict future 5G network service quality data exception, by the abnormal data predicted hair
It send to qos policy decision-making module；
Qos policy decision-making module 470: for marking and storing the abnormal data, abnormal results are reported and according to described
Abnormal data makes decisions decision, and the decision is driven to determine to execute.
The data acquisition module 410 specifically includes:
User terminal data acquisition unit: for obtaining hardware data, the software model version of user terminal, installation is answered
With, terminal location, moving direction, speed, CPU, memory source, the alarm log of consumption；
Access network data acquisition unit: for obtaining, base station distribution data, antenna channel mode, frequency spectrum are used, physics provides
Source virtual resource service condition, space interface signaling, alarm log data；
Core network data acquisition unit: for obtaining QoS of customer agreement, network is sliced resource and uses, core net letter
It enables, alarm log data；
Network KPI performance monitoring data acquisition unit: for obtaining network bandwidth, time delay, shake data；
The model training module 440 constructs the Supervised machine learning model, trained knot using decision Tree algorithms
Fruit is the tree being made of node and branch, and each non-leaf nodes indicates a category in the data set
Property, some value of branch, that is, attribute or value section, a classification in each leaf node, that is, data set indicate
Network service quality is abnormal or normal.
The qos policy decision-making module 470 ties the QoS exception monitoring device and the abnormal of the QoS predicting abnormality device
Fruit is marked, and new flag data is stored in the QoS data library, updates network service quality in the QoS data library
Data.
Data acquisition module 410, data processing module 420, model training module 440, QoS exception monitoring device 450, QoS
Predicting abnormality device 460, qos policy decision-making module 470 collectively form 5G network QoS machine learning engine, according to collected use
Family terminal QoS data, Access Network QoS data, core net QoS data and network KPI performance monitoring data, utilize the 5G network
QoS machine learning engine detects automatically and predicts network QoS exception, can be further formed network QoS management strategy, is 5G network
The quality of service guarantee of user also provides foundation for the network planning and network service quality optimization.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that each embodiment described in conjunction with the examples disclosed in this document
Module, unit and/or method and step can be realized with the combination of electronic hardware or computer software and electronic hardware.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations, although referring to before
Stating embodiment, invention is explained in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of 5G network service quality exception monitoring and prediction technique, which is characterized in that described method includes following steps:
S1, acquisition 5G network service quality data and network KPI performance monitoring data, the network service quality data include using
Family terminal data, access network data, core network data；
S2, the network service quality data are pre-processed and is marked；
S3, the network service quality data after the label are stored to QoS data library；
S4, using described in the QoS data library mark after network service quality data as data set, be built with supervision machine
Learning model is trained the Supervised machine learning model using the data set, obtain QoS exception monitoring device and
QoS predicting abnormality device；
S5, using the current 5G network service quality data of QoS exception monitoring device real-time monitoring, the abnormal data that will be monitored
It is sent to qos policy decision-making module；
S6, future 5G network service quality data exception, the abnormal data that will be predicted are predicted using the QoS predicting abnormality device
It is sent to qos policy decision-making module；
S7, the qos policy decision-making module mark and store the abnormal data, update QoS data library, report abnormal results are simultaneously
It is made decisions decision according to the abnormal data, the decision is driven to determine to execute.
2. 5G network service quality exception monitoring and prediction technique according to claim 1, which is characterized in that the step S1
In:
The user terminal data includes: the hardware data of user terminal, software model version, the application of installation, terminal location,
Moving direction, speed, CPU, memory source, the alarm log of consumption；
The access network data includes: base station distribution, antenna channel mode, frequency spectrum uses, physical resource virtual resource uses feelings
Condition, space interface signaling, alarm log；
The core network data includes: QoS of customer agreement, and network is sliced resource and uses, core network signalling, alarm log；
The network KPI performance monitoring data includes: network bandwidth, time delay, shake.
3. 5G network service quality exception monitoring and prediction technique according to claim 1, which is characterized in that the step S2
Detailed process are as follows:
To collect the network service quality data clean, unified format, in conjunction with the network KPI performance monitoring number
It is marked according to the network service quality data, the label includes normal and abnormal.
4. 5G network service quality exception monitoring and prediction technique according to claim 1, which is characterized in that the step S4
In, the Supervised machine learning model uses decision Tree algorithms.
5. 5G network service quality exception monitoring and prediction technique according to claim 1, which is characterized in that the step S6
In, the QoS predicting abnormality device is predicted according to current 5G network service quality data and history 5G network service quality data
The following 5G network service quality data exception saves in the history 5G network service quality data QoS data library
Web-based history quality of service data record.
6. 5G network service quality exception monitoring and prediction technique according to claim 1, which is characterized in that the step S7
In, the qos policy decision-making module marks and stores the abnormal data specifically:
Automatically the result of the QoS exception monitoring device and the QoS predicting abnormality device is marked, and by new flag data
It is stored in the QoS data library, updates network service quality data in the QoS data library.
7. a kind of 5G network service quality exception monitoring and forecasting system, which is characterized in that the system comprises:
Data acquisition module: for acquiring 5G network service quality data and network KPI performance monitoring data, the network service
Qualitative data includes user terminal QoS data, Access Network QoS data, core net QoS data；
Data processing module: for the network service quality data to be pre-processed and are marked；
Qos data memory module: for storing the network service quality data after the label；
Model training module: for using the network service quality data after being marked described in the QoS data library as there is supervision
The data set of machine learning model is built with supervision machine learning model, is trained to the Supervised machine learning model,
Obtain QoS exception monitoring device and QoS predicting abnormality device
QoS exception monitoring device: the current 5G network service quality data of real-time monitoring are used for, the abnormal data monitored is sent to
Qos policy decision-making module；
QoS predicting abnormality device: for according to the history in current 5G network service quality data and the Qos data memory module
5G network service quality data predict future 5G network service quality data exception, and the abnormal data predicted is sent to QoS
Policy decision module；
Qos policy decision-making module: for marking and storing the abnormal data, abnormal results are reported and according to the abnormal data
Make decisions decision, and the decision is driven to determine to execute.
8. 5G network service quality exception monitoring and forecasting system according to claim 7, which is characterized in that the data are adopted
Collection module specifically includes:
User terminal data acquisition unit: for obtaining hardware data, the software model version of user terminal, the application of installation,
Terminal location, moving direction, speed, CPU, memory source, the alarm log of consumption；
Access network data acquisition unit: it is used for obtaining base station distribution data, antenna channel mode, frequency spectrum, physical resource void
Quasi- resource service condition, space interface signaling, alarm log data；
Core network data acquisition unit: for obtaining QoS of customer agreement, network is sliced resource and uses, core network signalling,
Alarm log data；
Network KPI performance monitoring data acquisition unit: for obtaining network bandwidth, time delay, shake data.
9. 5G network service quality exception monitoring and forecasting system according to claim 7, which is characterized in that the model instruction
Practice module and constructs the Supervised machine learning model using decision Tree algorithms, it is training the result is that one by node and branch group
At tree, each non-leaf nodes indicates an attribute in the data set, branch, that is, attribute certain
A value is worth section, a classification in each leaf node, that is, data set, indicates that network service quality is abnormal or normal.
10. 5G network service quality exception monitoring and forecasting system according to claim 7, which is characterized in that the QoS plan
The abnormal results of the QoS exception monitoring device and the QoS predicting abnormality device are marked slightly decision-making module, and by new mark
Numeration updates network service quality data in the QoS data library according to the QoS data library is stored in.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910235520.3A CN109787846A (en) | 2019-03-27 | 2019-03-27 | A kind of 5G network service quality exception monitoring and prediction technique and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910235520.3A CN109787846A (en) | 2019-03-27 | 2019-03-27 | A kind of 5G network service quality exception monitoring and prediction technique and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109787846A true CN109787846A (en) | 2019-05-21 |
Family
ID=66491317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910235520.3A Pending CN109787846A (en) | 2019-03-27 | 2019-03-27 | A kind of 5G network service quality exception monitoring and prediction technique and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109787846A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111581045A (en) * | 2020-03-18 | 2020-08-25 | 平安科技（深圳）有限公司 | Database anomaly monitoring method and device, computer device and storage medium |
| CN112312427A (en) * | 2019-07-31 | 2021-02-02 | 中国移动通信集团广东有限公司 | Method for optimizing network quality and electronic equipment |
| CN113472590A (en) * | 2021-07-14 | 2021-10-01 | 南京邮电大学 | QoS configuration method based on machine learning in DDS |
| CN113541984A (en) * | 2020-04-14 | 2021-10-22 | 中国移动通信集团浙江有限公司 | Method, device and equipment for predicting experience of network slicing service |
| WO2021213247A1 (en) * | 2020-04-24 | 2021-10-28 | 华为技术有限公司 | Anomaly detection method and device |
| CN113747480A (en) * | 2020-05-28 | 2021-12-03 | 中国移动通信集团浙江有限公司 | Processing method and device for 5G slice fault and computing equipment |
| CN113825161A (en) * | 2020-06-18 | 2021-12-21 | 中国移动通信集团浙江有限公司 | 5G slice abnormity detection method and device based on deep self-coding neural network |
| CN114143163A (en) * | 2020-08-14 | 2022-03-04 | 中国移动通信集团浙江有限公司 | Slice false alarm identification method and device based on graph attention network |
| CN114615051A (en) * | 2022-03-09 | 2022-06-10 | 黄河水利职业技术学院 | Network security detection method and system |
| CN115996168A (en) * | 2021-10-20 | 2023-04-21 | 慧与发展有限责任合伙企业 | Supervised quality of service change derivation |
| CN116233871A (en) * | 2023-01-17 | 2023-06-06 | 广州爱浦路网络技术有限公司 | XR service enhancement method, computer device and storage medium |
| CN116302009A (en) * | 2023-05-19 | 2023-06-23 | 微网优联科技(成都)有限公司 | Software updating method and device based on wireless router |
| CN116528274A (en) * | 2023-07-05 | 2023-08-01 | 腾讯科技（深圳）有限公司 | Network quality regulation and control method and related equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8615575B2 (en) * | 2008-10-16 | 2013-12-24 | At&T Intellectual Property I, L.P. | Devices, methods, and computer-readable media for providing quality of service optimization via policy-based rearrangements |
| CN105119734A (en) * | 2015-07-15 | 2015-12-02 | 中国人民解放军防空兵学院 | Full network anomaly detection positioning method based on robust multivariate probability calibration model |
| CN106027288A (en) * | 2016-05-10 | 2016-10-12 | 华北电力大学 | Communication traffic prediction method for distribution line information monitoring service |
| CN106779069A (en) * | 2016-12-08 | 2017-05-31 | 国家电网公司 | A kind of abnormal electricity consumption detection method based on neutral net |
| CN109343995A (en) * | 2018-10-25 | 2019-02-15 | 金税信息技术服务股份有限公司 | Intelligent O&M analysis system based on multi-source heterogeneous data fusion, machine learning and customer service robot |
-
2019
- 2019-03-27 CN CN201910235520.3A patent/CN109787846A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8615575B2 (en) * | 2008-10-16 | 2013-12-24 | At&T Intellectual Property I, L.P. | Devices, methods, and computer-readable media for providing quality of service optimization via policy-based rearrangements |
| CN105119734A (en) * | 2015-07-15 | 2015-12-02 | 中国人民解放军防空兵学院 | Full network anomaly detection positioning method based on robust multivariate probability calibration model |
| CN106027288A (en) * | 2016-05-10 | 2016-10-12 | 华北电力大学 | Communication traffic prediction method for distribution line information monitoring service |
| CN106779069A (en) * | 2016-12-08 | 2017-05-31 | 国家电网公司 | A kind of abnormal electricity consumption detection method based on neutral net |
| CN109343995A (en) * | 2018-10-25 | 2019-02-15 | 金税信息技术服务股份有限公司 | Intelligent O&M analysis system based on multi-source heterogeneous data fusion, machine learning and customer service robot |
Non-Patent Citations (2)
| Title |
|---|
| GUOSHENG ZHU，JUN ZANZ, YANG YANG，AND XIAOYUN QI: "A Supervised Learning Based QoS Assurance Architecture for 5G Networks", 《IEEE ACCESS》 * |
| 陈胜等: "基于C5.0决策树的NAT设备检测方法", 《计算机科学》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112312427B (en) * | 2019-07-31 | 2023-09-05 | 中国移动通信集团广东有限公司 | Method for optimizing network quality and electronic equipment |
| CN112312427A (en) * | 2019-07-31 | 2021-02-02 | 中国移动通信集团广东有限公司 | Method for optimizing network quality and electronic equipment |
| CN111581045B (en) * | 2020-03-18 | 2024-05-28 | 平安科技（深圳）有限公司 | Database anomaly monitoring method, device, computer device and storage medium |
| CN111581045A (en) * | 2020-03-18 | 2020-08-25 | 平安科技（深圳）有限公司 | Database anomaly monitoring method and device, computer device and storage medium |
| CN113541984A (en) * | 2020-04-14 | 2021-10-22 | 中国移动通信集团浙江有限公司 | Method, device and equipment for predicting experience of network slicing service |
| WO2021213247A1 (en) * | 2020-04-24 | 2021-10-28 | 华为技术有限公司 | Anomaly detection method and device |
| CN113747480A (en) * | 2020-05-28 | 2021-12-03 | 中国移动通信集团浙江有限公司 | Processing method and device for 5G slice fault and computing equipment |
| CN113747480B (en) * | 2020-05-28 | 2024-05-07 | 中国移动通信集团浙江有限公司 | Processing method and device for 5G slice faults and computing equipment |
| CN113825161B (en) * | 2020-06-18 | 2023-08-04 | 中国移动通信集团浙江有限公司 | 5G slice anomaly detection method and device based on depth self-coding neural network |
| CN113825161A (en) * | 2020-06-18 | 2021-12-21 | 中国移动通信集团浙江有限公司 | 5G slice abnormity detection method and device based on deep self-coding neural network |
| CN114143163B (en) * | 2020-08-14 | 2023-11-14 | 中国移动通信集团浙江有限公司 | Slice false alarm identification method and device based on graph attention network |
| CN114143163A (en) * | 2020-08-14 | 2022-03-04 | 中国移动通信集团浙江有限公司 | Slice false alarm identification method and device based on graph attention network |
| CN113472590A (en) * | 2021-07-14 | 2021-10-01 | 南京邮电大学 | QoS configuration method based on machine learning in DDS |
| CN115996168A (en) * | 2021-10-20 | 2023-04-21 | 慧与发展有限责任合伙企业 | Supervised quality of service change derivation |
| CN115996168B (en) * | 2021-10-20 | 2024-04-05 | 慧与发展有限责任合伙企业 | Supervised quality of service change derivation |
| CN114615051A (en) * | 2022-03-09 | 2022-06-10 | 黄河水利职业技术学院 | Network security detection method and system |
| CN116233871A (en) * | 2023-01-17 | 2023-06-06 | 广州爱浦路网络技术有限公司 | XR service enhancement method, computer device and storage medium |
| CN116233871B (en) * | 2023-01-17 | 2023-12-15 | 广州爱浦路网络技术有限公司 | XR service enhancement method, computer device and storage medium |
| CN116302009B (en) * | 2023-05-19 | 2023-08-08 | 微网优联科技(成都)有限公司 | Software updating method and device based on wireless router |
| CN116302009A (en) * | 2023-05-19 | 2023-06-23 | 微网优联科技(成都)有限公司 | Software updating method and device based on wireless router |
| CN116528274B (en) * | 2023-07-05 | 2023-09-22 | 腾讯科技（深圳）有限公司 | Network quality regulation and control method and related equipment |
| CN116528274A (en) * | 2023-07-05 | 2023-08-01 | 腾讯科技（深圳）有限公司 | Network quality regulation and control method and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109787846A (en) | A kind of 5G network service quality exception monitoring and prediction technique and system | |
| WO2021052394A1 (en) | Model training method, apparatus, and system | |
| EP3968243A1 (en) | Method and apparatus for realizing model training, and computer storage medium | |
| CN101826090A (en) | WEB public opinion trend forecasting method based on optimal model | |
| CN105071983A (en) | Abnormal load detection method for cloud calculation on-line business | |
| KR101542235B1 (en) | Method for mining of real-time data | |
| CN115225536B (en) | Virtual machine abnormality detection method and system based on unsupervised learning | |
| CN103678004A (en) | Host load prediction method based on unsupervised feature learning | |
| CN108470022A (en) | A kind of intelligent work order quality detecting method based on operation management | |
| CN112884159B (en) | Model updating system, model updating method and related equipment | |
| Shi et al. | Anomaly detection for key performance indicators through machine learning | |
| CN105471647A (en) | Power communication network fault positioning method | |
| US11197175B2 (en) | Forcasting time series data | |
| Gholami et al. | Maintenance scheduling using data mining techniques and time series models | |
| Wang et al. | Predicting the effectiveness of supplement time on delay recoveries: a support vector regression approach | |
| CN105761093A (en) | Knowledge-space-based behavior result evaluation method and device | |
| CN113825165B (en) | 5G slice network congestion early warning method and device based on time diagram network | |
| Wang et al. | A new time series prediction method based on complex network theory | |
| Zhang et al. | Intelligent travelling visitor estimation model with big data mining | |
| CN116522213A (en) | Service state level classification and classification model training method and electronic equipment | |
| Sayed-Mouchaweh | Learning from Data Streams in Evolving Environments: Methods and Applications | |
| Costa et al. | A three level sensor ranking method based on active perception | |
| Yang | Intelligent informatization early warning analysis of agricultural economy based on support vector sequential regression model | |
| Wang et al. | Intelligent Network operation and maintenance based on deep learning technology | |
| Sammour et al. | A validation measure for computational scheduler activity-based transportation models based on sequence alignment methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190521 |