CN114423049A - Perception prediction method and device, electronic equipment and storage medium - Google Patents
Perception prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114423049A CN114423049A CN202111422289.2A CN202111422289A CN114423049A CN 114423049 A CN114423049 A CN 114423049A CN 202111422289 A CN202111422289 A CN 202111422289A CN 114423049 A CN114423049 A CN 114423049A
- Authority
- CN
- China
- Prior art keywords
- air interface
- data
- sample
- service data
- interface data
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5003—Managing SLA; Interaction between SLA and QoS
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a perception prediction method, a perception prediction device, electronic equipment and a storage medium, which are applied to a 5G network, wherein the method comprises the following steps: acquiring air interface data and service data in real time, determining a characteristic vector of the air interface data and a characteristic vector of the service data, inputting the air interface data and the service data into a pre-trained perception prediction model, obtaining a QoE index prediction value, determining a target control strategy according to the QoE index prediction value and a preset QoE index threshold value, and issuing the target control strategy; the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data. The perception prediction method provided by the invention can be applied to various application scenes, parameters in a control strategy are adaptively adjusted, the high speed and low delay of a wireless network are ensured in real time, and the user experience is improved.
Description
Technical Field
The present invention relates to the field of wireless network technologies, and in particular, to a perception prediction method and apparatus, an electronic device, and a storage medium.
Background
With the continuous development of intelligent processing technology, it is crucial for users to improve the quality of service of wireless networks.
Currently, in 3GPP (3rd Generation Partnership Project) protocol, a 5G (5-Generation mobile communication technology) network is defined. In order to meet the requirements of 5G users, in the prior art, a semi-static QoS configuration policy is adopted to adapt to fluctuation change of air interface transmission data, but this processing method cannot adaptively adjust QoS configuration parameters along with fluctuation change of air interface data, and needs to adapt to corresponding QoS configuration policies for different application scenarios, so that the application cost is high, and high speed and low delay of a wireless network cannot be guaranteed in real time, which results in poor user experience.
Disclosure of Invention
The invention provides a perception prediction method, a perception prediction device, electronic equipment and a storage medium, which are used for solving the technical problems that the high speed of a wireless network cannot be ensured in real time and the user experience is poor due to the fact that QoS configuration parameters cannot be adjusted and changed in real time in the prior art.
In a first aspect, the present invention provides a perceptual prediction method applied in a 5G network, including:
acquiring air interface data and service data in real time;
determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value;
determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issuing the target control strategy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
Further, according to the perceptual prediction method provided by the present invention, determining a target control policy according to the QoE metric prediction value and a preset QoE metric threshold value includes:
and under the condition that the QoE index predicted value is greater than or equal to a preset QoE index threshold value, determining an initial control strategy as a target control strategy without adjustment.
Further, according to the perceptual prediction method provided by the present invention, determining a target control policy according to the QoE metric prediction value and a preset QoE metric threshold value further includes:
and under the condition that the QoE index predicted value is smaller than a preset QoE index threshold value, adjusting the initial control strategy, and determining the adjusted control strategy as a target control strategy.
Further, according to the perceptual prediction method provided by the present invention, the acquiring of the air interface data and the service data in real time includes:
and acquiring air interface data and service data in real time through the E2 interface data acquisition service.
Further, according to the perceptual prediction method provided by the present invention, the determining the feature vector of the air interface data and the feature vector of the service data includes:
acquiring the air interface data and the service data from a message queue, and cleaning the air interface data and the service data;
and carrying out vector processing on the cleaned air interface data and the service data to obtain a characteristic vector of the air interface data and a characteristic vector of the service data.
Further, according to the perceptual prediction method provided by the present invention, before the acquiring the air interface data and the service data in real time, the method includes:
acquiring sample air interface data, marking information of the sample air interface data, sample service data and marking information of the sample service data;
carrying out vector processing on the sample air interface data and the sample service data to obtain a characteristic vector of the sample air interface data and a characteristic vector of the sample service data;
and training a perception prediction model based on the characteristic vector of the sample air interface data, the labeling information of the sample air interface data, the characteristic vector of the sample service data and the labeling information of the sample service data.
Further, according to the perceptual prediction method provided by the present invention, the training of a perceptual prediction model based on the feature vector of the sample air interface data, the label information of the sample air interface data, the feature vector of the sample service data, and the label information of the sample service data includes:
step S1, predicting the characteristic vector of the sample air interface data and the characteristic vector of the sample service data by using the emotion classification model to be trained to obtain a QoE prediction result;
step S2, comparing the QoE prediction result with the labeling information of the sample air interface data and the labeling information of the sample service data, judging whether a model training termination condition is met, adjusting the perception prediction model to be trained when the model training termination condition is not met, and executing step S1 again by using the adjusted perception prediction model; and when the model training termination condition is met, obtaining a trained perception prediction model.
In a second aspect, the present invention further provides a perceptual prediction apparatus, applied in a 5G network, including:
the acquisition module is used for acquiring air interface data and service data in real time;
the determining and inputting module is used for determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value;
the determining module is used for determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value and issuing the target control strategy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the perceptual prediction method as described in any one of the above.
In a fourth aspect, the invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the perceptual prediction method as defined in any one of the above.
In a fifth aspect, the present invention also provides a computer program product comprising computer executable instructions for implementing the steps of the perceptual prediction method as defined in any one of the above.
The invention provides a perception prediction method, a perception prediction device, electronic equipment and a storage medium, which are applied to a 5G network, empty data and service data are collected in real time, a characteristic vector of the empty data and a characteristic vector of the service data are determined, the empty data and the service data are input into a perception prediction model trained in advance, a QoE index prediction value is obtained, a target control strategy is determined according to the QoE index prediction value and a preset QoE index threshold value, and the target control strategy is issued; the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data. The perception prediction method provided by the invention can be applied to various application scenes, parameters in a control strategy are adaptively adjusted, the high speed and low delay of a wireless network are ensured in real time, and the user experience is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a perceptual prediction method provided by the present invention;
FIG. 2 is a flow chart of the overall processing procedure of a perception prediction method provided by the present invention;
FIG. 3 is a schematic structural diagram of a complaint handling device provided by the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flowchart of a perceptual prediction method provided by the present invention, and as shown in fig. 1, the perceptual prediction method provided by the present invention is applied in a 5G wireless network, and specifically includes the following steps:
step 101: and acquiring air interface data and service data in real time.
In this embodiment, the air interface data and the service data of the user need to be acquired in real time, and the real-time performance of data processing is ensured by performing associated analysis on the acquired air interface data and the service data at the same time. The air interface data is obtained from an air interface data packet, and the air interface data packet refers to a high-frequency data resource used for transmission between a mobile phone or other wireless internet access tools and a base station. Only if the effective air interface data resources are available, the uploading and issuing of the information can be realized. The service data refers to data generated by operation data of a user, such as data generated by operations of watching videos, downloading pictures, chatting videos, and the like.
It should be noted that, in this embodiment, the module collects air interface data and service data in real time through an E2 interface data collection service module, and the module collects the data in real time after receiving a collection task issued by an engine device management service, and sends the collected data to a KAFKA message queue and a relational database (Redis) in a database management system, respectively, where KAFKA is a distributed, high-throughput, high-scalability message queue system, and a Redis (remote Dictionary server), that is, a remote Dictionary service, which is an open-source database written in ANSI C language, supporting a network, a memory-based and Key-based persistent log-type and Value-Value database.
Step 102: and determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value.
In this embodiment, vectorization processing needs to be performed on the air interface data and the service data obtained in step 101, to determine a feature vector of the air interface data and a feature vector of the service data, and to input the obtained feature vectors into a pre-trained perceptual prediction model, so as to obtain a QoE index prediction value.
It should be noted that QoE (Quality of Experience) refers to the subjective feeling of the user on the Quality and performance of the device, network and system, application or service, and a QoE index includes multiple index data, where multiple index values include, but are not limited to, the following: the method includes the steps of receiving rate, dropping rate, average utilization rate of a transmission Resource Block (PRB), co-channel interference and HTTP download rate, where the index values are collectively referred to as Key Performance Indicators (KPIs), and if one of the index values is smaller than a preset index threshold, a set control strategy needs to be adjusted to re-determine a new control strategy.
Step 103: determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issuing the target control strategy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
In this embodiment, the QoE index prediction value obtained through the perceptual prediction model in step 102 is compared with a preset QoE index threshold, and when any one index value in the QoE index prediction value is smaller than the preset QoE index threshold, the target control policy is re-determined and is issued to the base station E2 interface control policy, so as to optimize the base station QoE control policy. The control strategy is a strategy for controlling the state of the user connection network, and a control method for ensuring the user connection network to keep high speed and low delay in real time by adjusting specific parameters.
It should be noted that, in the present embodiment, the target control policy refers to QoS parameter configuration, where QoS (Quality of Service) refers to a network being capable of providing better Service capability for specified network communication by using various basic technologies.
It should be noted that the perceptual prediction model is obtained by training based on the sample air interface data and the labeling information of the sample air interface data, the sample service data and the labeling information of the sample service data, and is obtained by training based on a Long Short-Term Memory network (LSTM) and a Deep Neural network (DNN for Short).
The invention provides a perception prediction method, which is applied to a 5G network, collects air interface data and service data in real time, determines a characteristic vector of the air interface data and a characteristic vector of the service data, inputs the air interface data and the characteristic vector of the service data into a perception prediction model trained in advance, obtains a QoE index predicted value, determines a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issues the target control strategy; the perception prediction model is obtained by training based on the sample air interface data and the labeling information of the sample air interface data, the sample service data and the labeling information of the sample service data. The perception prediction method provided by the invention can be applied to various application scenes, parameters in a control strategy are adaptively adjusted, the high speed and low delay of a wireless network are ensured in real time, and the user experience is improved.
Based on any one of the foregoing embodiments, in this embodiment, the determining a target control policy according to the QoE metric predicted value and a preset QoE metric threshold includes:
and under the condition that the QoE index predicted values are all larger than or equal to a preset QoE index threshold value, determining an initial control strategy as a target control strategy without adjustment.
In this embodiment, when the QoE index prediction values obtained through model prediction are all greater than or equal to the preset QoE index threshold, the initial control strategy is determined as the target strategy without being adjusted. The control strategy is a strategy for controlling the state of the user connection network, and a control method for ensuring the user connection network to keep high speed and low delay in real time by adjusting specific parameters.
For example, if the preset QoE index threshold includes a call completing rate of 20%, an average utilization rate of transmission resource blocks of 80%, and an HTTP download rate of 300M/s, it is predicted through a model that QoE index predicted values corresponding to a network 1 connected to the user 1 are respectively a call completing rate of 30%, an average utilization rate of transmission resource blocks of 90%, and an HTTP download rate of 400M/s, and the predicted QoE index predicted values of the user 1 in the next time period are all greater than the preset QoE index threshold, which indicates that the network state is normal, the initial control policy is not adjusted, and the initial control policy is directly determined as a target policy for issuing processing.
According to the perception prediction method provided by the invention, the corresponding control strategy can be determined according to the relationship between the QoE index predicted value and the QoE index threshold value, the high-speed and low-delay use state of the 5G network of the user is ensured, and the user experience is improved.
Based on any one of the foregoing embodiments, in this embodiment, the determining a target control policy according to the QoE metric predicted value and a preset QoE metric threshold further includes:
and under the condition that the QoE index predicted value is smaller than a preset QoE index threshold value, adjusting the initial control strategy, and determining the adjusted control strategy as a target control strategy.
In this embodiment, when the predicted value of any one of the QoE index predicted values obtained through model prediction is smaller than a preset QoE index threshold, the initial control strategy is adjusted, and the adjusted control strategy is determined as the target strategy.
For example, if the preset QoE index threshold includes a call completing rate of 20%, an average utilization rate of transmission resource blocks of 80%, and an HTTP download rate of 300M/s, it is predicted through a model that QoE index predicted values corresponding to a network 1 connected to the user 1 are respectively a call completing rate of 10%, an average utilization rate of transmission resource blocks of 90%, and an HTTP download rate of 400M/s, and the predicted call completing rate of the user 1 in the QoE index predicted values in the following time period is lower than the preset threshold, indicating that the network state is abnormal, adjusting corresponding parameters in an initial control strategy, and then determining the adjusted control strategy as a target strategy for issuing processing.
According to the perception prediction method provided by the invention, the corresponding control strategy can be determined according to the relationship between the QoE index predicted value and the QoE index threshold value, the high-speed and low-delay use state of the 5G network of the user is ensured, and the user experience is improved.
Based on any one of the above embodiments, in this embodiment, the acquiring the air interface data and the service data in real time includes:
and acquiring air interface data and service data in real time through the E2 interface data acquisition service.
In this embodiment, the air interface data and the service data are collected in real time by an E2 interface data collection service, and the air interface data and the service data are collected in real time by an E2 interface data collection service when receiving a collection task issued by an engine device management service, where the base station E2 interface information and data collected in real time are respectively sent to a KAFKA information queue and a Redis database in a database management system to store original data.
According to the perception prediction method provided by the invention, the air interface data and the service data are collected in real time through the E2 interface data collection service, so that the dynamic data processing is realized, and the high speed and the low time delay of the network are ensured.
Based on any of the foregoing embodiments, in this embodiment, the determining the feature vector of the air interface data and the feature vector of the service data includes:
acquiring the air interface data and the service data from a message queue, and cleaning the air interface data and the service data;
and carrying out vector processing on the cleaned air interface data and the service data to obtain a characteristic vector of the air interface data and a characteristic vector of the service data.
In this embodiment, corresponding data is extracted from the KAFKA message queue in which air interface data and service data are stored in advance, the obtained air interface data and service data are cleaned, vectorization processing is performed on the cleaned data, and a feature vector of the air interface data and a feature vector of the service data are determined and used in subsequent prediction processing. It should be noted that both the cleaning and the vector processing are performed by the well-established processing techniques in the prior art, and are not limited herein.
It should be noted that, in this embodiment, the processed data needs to be sent to a Redis database in a database management system for storage, and then the data is synchronously updated to MySQL data for storage, so as to be used for subsequent data retrieval, where MySQL is a better relational database.
According to the perception prediction method provided by the invention, the obtained data can be applied to the prediction of the perception prediction model by cleaning and vectorizing the data obtained from the message queue, so that data support is provided for determining a target control strategy, and the processing efficiency is improved.
Based on any one of the above embodiments, in this embodiment, before the acquiring the air interface data and the service data in real time, the method includes:
acquiring sample air interface data, marking information of the sample air interface data, sample service data and marking information of the sample service data;
carrying out vector processing on the sample air interface data and the sample service data to obtain a characteristic vector of the sample air interface data and a characteristic vector of the sample service data;
and training a perception prediction model based on the characteristic vector of the sample air interface data, the labeling information of the sample air interface data, the characteristic vector of the sample service data and the labeling information of the sample service data.
In this embodiment, vector processing needs to be performed on the obtained sample air interface data and sample service data, so as to obtain a feature vector of the sample air interface data and a feature vector of the sample service data, and then a perceptual prediction model is trained based on the feature vector of the sample air interface data and the feature vector of the sample service data, and respective corresponding labeling information. If the labeling information 1 corresponding to the sample air interface data 1 is normal and the labeling information 1 corresponding to the sample service data 1 is abnormal, it should be noted that the specific labeling mode of the labeling information may be a labeling mode, and specifically may be set according to the actual needs of the user, which is not specifically limited herein.
It should be noted that, in this embodiment, multiple deep learning algorithms are introduced into the perceptual prediction model by accessing XAPP management to engine device management service, XAPP (third party APP) is accessed through an API interface, a permutation and combination of multiple perceptual prediction capabilities are realized, an optimal QoE index prediction value is obtained, accuracy of data is ensured, and an optimal target control policy is determined.
According to the perception prediction method provided by the invention, vector processing is carried out on the acquired sample air interface data and the sample service data, and then the perception prediction model is trained according to the acquired characteristic vector and the marking information.
Based on any one of the foregoing embodiments, in this embodiment, the training a perceptual prediction model based on the feature vector of the sample air interface data, the label information of the sample air interface data, the feature vector of the sample service data, and the label information of the sample service data includes:
step S1, predicting the characteristic vector of the sample air interface data and the characteristic vector of the sample service data by using the emotion classification model to be trained to obtain a QoE prediction result;
step S2, comparing the QoE prediction result with the labeling information of the sample air interface data and the labeling information of the sample service data, judging whether a model training termination condition is met, adjusting the perception prediction model to be trained when the model training termination condition is not met, and executing step S1 again by using the adjusted perception prediction model; and when the model training termination condition is met, obtaining a trained perception prediction model.
In this embodiment, a perceptual prediction model is trained according to the obtained feature vector of the sample air interface data and the feature vector of the sample service data, when the obtained QoE prediction result is consistent with the label information, the prediction result is correct, and only when the result of multiple times of training meets the training termination condition, the training is stopped to obtain the perceptual prediction model; and if the model training termination condition is not met, continuing training. Wherein, the termination condition of the model training may be that the accuracy is 90%, and the training is stopped only when the accuracy of the training is more than 90%. The model training termination condition may be specifically set according to actual needs, and is not specifically limited herein.
According to the perception prediction method provided by the invention, when the model training termination condition is met, the model training is stopped, the trained perception prediction model is obtained, the prediction accuracy of the perception prediction model is ensured, and the data processing efficiency is improved.
Based on any of the above embodiments, in this embodiment, as shown in fig. 2, air interface data (mobile network data) and service data (user platform data) are collected in real time, then the collected data are preprocessed, cleaned, and vector-processed, a feature vector of the air interface data and a feature vector of the service data obtained after preprocessing are input into a pre-trained perceptual prediction model, a predicted QoE index value is obtained, a target control policy (QoS parameter configuration) is determined according to a relationship between the QoE index predicted value and a QoE index threshold, and the target control policy is issued.
It should be noted that, a cell KPI (Key performance Indicator) value is a QoE Indicator threshold value provided by the present invention, a user KQI Indicator value is monitored in real time as each Indicator value in a QoE Indicator prediction value, and a control strategy is adjusted according to a relationship between the cell KPI Indicator value and the user KQI Indicator value to determine a target control strategy.
It should be noted that, in this embodiment, the engine device management service is further accessed to the third-party application APP through the API interface, an open software ecological framework is constructed, access of application software that is in accordance with a certain specific application scenario is provided, and different business requirements of users in different industries are met.
Fig. 3 is a perceptual prediction apparatus applied in a 5G network, and the apparatus includes:
the acquisition module 301 is used for acquiring air interface data and service data in real time;
a determining and inputting module 302, configured to determine a feature vector of the air interface data and a feature vector of the service data, input the feature vectors into a pre-trained perceptual prediction model, and obtain a QoE index prediction value;
a determining module 303, configured to determine a target control policy according to the QoE index prediction value and a preset QoE index threshold, and issue the target control policy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
The invention provides a perception prediction device which is applied to a 5G network, empty port data and service data need to be collected in real time, a characteristic vector of the empty port data and a characteristic vector of the service data are determined, the empty port data and the characteristic vectors of the service data are input into a perception prediction model trained in advance, a QoE index prediction value is obtained, a target control strategy is determined according to the QoE index prediction value and a preset QoE index threshold value, and the target control strategy is issued; the perception prediction model is obtained by training based on the sample air interface data and the labeling information of the sample air interface data, the sample service data and the labeling information of the sample service data. The perception prediction device provided by the invention can be applied to various application scenes, adaptively adjusts parameters in a control strategy, ensures high speed and low delay of a wireless network in real time, and improves user experience.
Since the principle of the apparatus according to the embodiment of the present invention is the same as that of the method according to the above embodiment, further details are not described herein for further explanation.
Fig. 4 is a schematic structural diagram of an electronic device provided in an embodiment of the present invention, and as shown in fig. 4, the present invention provides an electronic device, including: a processor (processor)401, a memory (memory)402, and a bus 403;
the processor 401 and the memory 402 complete communication with each other through the bus 403;
Furthermore, the logic instructions in the memory 402 may be implemented in software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer being capable of executing the perceptual prediction method provided by the above methods, the method being applied in a 5G network, and including: acquiring air interface data and service data in real time; determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value; determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issuing the target control strategy; the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to perform the perceptual prediction methods provided above, the method being applied in a 5G network, and comprising: acquiring air interface data and service data in real time; determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value; determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issuing the target control strategy; the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A perception prediction method is applied to a 5G network and comprises the following steps:
acquiring air interface data and service data in real time;
determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value;
determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value, and issuing the target control strategy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
2. The perceptual prediction method of claim 1, wherein the determining a target control policy according to the QoE metric prediction value and a preset QoE metric threshold value comprises:
and under the condition that the QoE index predicted value is greater than or equal to a preset QoE index threshold value, determining an initial control strategy as a target control strategy without adjustment.
3. The perceptual prediction method of claim 2, wherein the determining a target control policy according to the QoE metric prediction value and a preset QoE metric threshold value further comprises:
and under the condition that the QoE index predicted value is smaller than a preset QoE index threshold value, adjusting the initial control strategy, and determining the adjusted control strategy as a target control strategy.
4. The perceptual prediction method of claim 1 wherein the collecting of air interface data and traffic data in real-time comprises:
and acquiring air interface data and service data in real time through the E2 interface data acquisition service.
5. The perceptual prediction method of claim 1, wherein the determining the empty port data eigenvector and the service data eigenvector comprises:
acquiring the air interface data and the service data from a message queue, and cleaning the air interface data and the service data;
and carrying out vector processing on the cleaned air interface data and the service data to obtain a characteristic vector of the air interface data and a characteristic vector of the service data.
6. The perceptual prediction method of claim 1, comprising, prior to the acquiring the air interface data and the traffic data in real-time:
acquiring sample air interface data, marking information of the sample air interface data, sample service data and marking information of the sample service data;
carrying out vector processing on the sample air interface data and the sample service data to obtain a characteristic vector of the sample air interface data and a characteristic vector of the sample service data;
and training a perception prediction model based on the characteristic vector of the sample air interface data, the labeling information of the sample air interface data, the characteristic vector of the sample service data and the labeling information of the sample service data.
7. The perceptual prediction method of claim 6, wherein training a perceptual prediction model based on the feature vector of the sample air interface data and the label information of the sample air interface data, the feature vector of the sample service data and the label information of the sample service data comprises:
step S1, predicting the characteristic vector of the sample air interface data and the characteristic vector of the sample service data by using the emotion classification model to be trained to obtain a QoE prediction result;
step S2, comparing the QoE prediction result with the labeling information of the sample air interface data and the labeling information of the sample service data, judging whether a model training termination condition is met, adjusting the perception prediction model to be trained when the model training termination condition is not met, and executing step S1 again by using the adjusted perception prediction model; and when the model training termination condition is met, obtaining a trained perception prediction model.
8. A perception prediction device is applied to a 5G network and comprises the following components:
the acquisition module is used for acquiring air interface data and service data in real time;
the determining and inputting module is used for determining the characteristic vector of the air interface data and the characteristic vector of the service data, inputting the characteristic vectors into a pre-trained perception prediction model, and obtaining a QoE index prediction value;
the determining module is used for determining a target control strategy according to the QoE index predicted value and a preset QoE index threshold value and issuing the target control strategy;
the perception prediction model is obtained by training based on sample air interface data and the labeling information of the sample air interface data, sample service data and the labeling information of the sample service data.
9. An electronic device, comprising:
a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the steps of the perceptual prediction method as defined in any one of claims 1 to 7.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the steps of the perceptual prediction method as defined in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111422289.2A CN114423049A (en) | 2021-11-26 | 2021-11-26 | Perception prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111422289.2A CN114423049A (en) | 2021-11-26 | 2021-11-26 | Perception prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114423049A true CN114423049A (en) | 2022-04-29 |
Family
ID=81266492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111422289.2A Pending CN114423049A (en) | 2021-11-26 | 2021-11-26 | Perception prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114423049A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024036953A1 (en) * | 2022-08-18 | 2024-02-22 | 中兴通讯股份有限公司 | Service experience guarantee method, electronic device and computer-readable storage medium |
-
2021
- 2021-11-26 CN CN202111422289.2A patent/CN114423049A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024036953A1 (en) * | 2022-08-18 | 2024-02-22 | 中兴通讯股份有限公司 | Service experience guarantee method, electronic device and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240089173A1 (en) | Multi-access edge computing based visibility network | |
| US20210027173A1 (en) | Indicator determining method and related device | |
| CN109302719B (en) | LTE cell capacity prediction analysis method and device | |
| US20230146912A1 (en) | Method, Apparatus, and Computing Device for Constructing Prediction Model, and Storage Medium | |
| CN108989880B (en) | Code rate self-adaptive switching method and system | |
| CN113543160B (en) | 5G slice resource allocation method, device, computing equipment and computer storage medium | |
| WO2019206100A1 (en) | Feature engineering programming method and apparatus | |
| CN112312427A (en) | Method for optimizing network quality and electronic equipment | |
| CN105830415A (en) | Methods, radio communication device and base station device for managing a media stream | |
| US11558769B2 (en) | Estimating apparatus, system, method, and computer-readable medium, and learning apparatus, method, and computer-readable medium | |
| Xu et al. | Modeling buffer starvations of video streaming in cellular networks with large-scale measurement of user behavior | |
| CN118102318B (en) | Data transmission system based on 5G technology | |
| CN114423049A (en) | Perception prediction method and device, electronic equipment and storage medium | |
| EP3073736B1 (en) | Method and device for measuring quality of experience of mobile video service | |
| EP2930617A1 (en) | Resource management method and device | |
| CN111314489B (en) | Method, server and device for identifying type of access network | |
| CN115086940A (en) | QoS (quality of service) adjusting method, system and device based on 5G and storage medium | |
| Raca et al. | Device-based Cellular Throughput Prediction for Video Streaming: Lessons from a Real-World Evaluation | |
| CN115484624A (en) | Data processing method, architecture, electronic device and storage medium | |
| EP2813033B1 (en) | Technique for projecting network load in a communication network | |
| CN117424860A (en) | Service function chain SFC resource allocation method and device | |
| CN113179218B (en) | Model training method, network congestion control method, device and related products | |
| CN107027130B (en) | Information acquisition method and device | |
| CN109327329B (en) | Data model updating method and device | |
| WO2023179709A1 (en) | Information processing method and apparatus, communication device, and readable storage medium |
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 |