CN108934016B

CN108934016B - Method and device for dividing cell scene categories, computer equipment and storage medium

Info

Publication number: CN108934016B
Application number: CN201810723557.6A
Authority: CN
Inventors: 刘津羽; 董陈小玉; 莫景画; 唐艺龙
Original assignee: Guangdong Haige Icreate Technology Co ltd
Current assignee: Guangdong Haige Icreate Technology Co ltd
Priority date: 2018-07-04
Filing date: 2018-07-04
Publication date: 2021-10-29
Anticipated expiration: 2038-07-04
Also published as: CN108934016A

Abstract

The application relates to a method, a system, computer equipment and a storage medium for dividing cell scene categories. The method comprises the following steps: acquiring first key performance index data of a cell to be divided, and acquiring first communication behavior characteristic data of the cell to be divided from the first key performance index data; performing principal component dimensionality reduction on the relevant data of each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of the cell to be divided; generating a first communication behavior oscillogram by utilizing each first communication behavior characteristic value and combining time granularity; and acquiring the scene category of the cell to be divided according to the first communication behavior oscillogram. According to the method, the communication behavior oscillogram is used as a dividing basis of the scene categories of the cells, so that the accuracy of the division of the scene categories of the cells is greatly improved, and an important basis is provided for customizing a planning construction scheme and determining a network optimization strategy.

Description

Method and device for dividing cell scene categories, computer equipment and storage medium

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method and an apparatus for dividing cell scene categories, a computer device, and a storage medium.

Background

With the continuous development of mobile communication networks, scene division of different cells is an important basis for customizing a planning construction scheme and determining a network optimization strategy.

The traditional cell scene division of the mobile communication network mainly depends on a network planning engineer to carry out artificial division according to the geographic environment, the coverage factor, the service characteristics and the like of a cell, the dependence on the personal experience of the network planning engineer is large, but the division of the coarse granularity of the cell scene according to the geographic environment, the coverage factor and the service characteristics of the cell often cannot accurately and effectively identify the network characteristics of the cell, so that the accuracy of the cell scene division is low, and the requirements of network fine modulation optimization and planning construction are difficult to meet.

Disclosure of Invention

Based on this, it is necessary to provide a method, an apparatus, a computer device and a storage medium for dividing cell scene categories, aiming at the technical problems that the traditional technology often cannot accurately and effectively identify cell network features, so that the accuracy of cell scene division is low, and network optimization and construction requirements are difficult to meet.

A method for dividing cell scene categories comprises the following steps:

acquiring first key performance index data of a cell to be divided, and acquiring first communication behavior characteristic data of the cell to be divided from the first key performance index data;

performing principal component dimensionality reduction on the relevant data of each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of the cell to be divided;

generating a first communication behavior oscillogram by utilizing each first communication behavior characteristic value and combining time granularity;

and acquiring the scene category of the cell to be divided according to the first communication behavior oscillogram.

An apparatus for dividing cell scene categories, comprising:

the communication behavior characteristic data acquisition module is used for acquiring first key performance index data of a cell to be divided and acquiring the first communication behavior characteristic data of the cell to be divided from the first key performance index data;

the communication behavior characteristic value acquisition module is used for performing principal component dimension reduction on the first communication behavior related data by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of the cell to be divided;

the communication behavior oscillogram acquisition module is used for generating a first communication behavior oscillogram by utilizing each first communication behavior characteristic value and combining time granularity;

and the scene category dividing module is used for acquiring the scene category of the cell to be divided according to the first communication behavior oscillogram.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

According to the method, the device, the computer equipment and the storage medium for dividing the cell scene categories, the key performance index data of the cell are obtained, the communication behavior characteristic value is obtained through dimensionality reduction of the field data related to the communication behavior characteristic in the key performance index data, then the communication behavior oscillogram of the cell is obtained according to the communication behavior characteristic value, the key performance index data of the cell is converted into a visual communication behavior oscillogram, the communication behavior oscillogram is used as a dividing basis of the scene categories of the cell, accurate division of the cell scene categories is achieved, and an important basis is provided for customizing a planning construction scheme and determining a network optimization strategy.

Drawings

Fig. 1 is an application environment diagram of a method for dividing cell scene categories according to an embodiment of the present invention;

fig. 2 is a flowchart of a method for dividing cell scene categories according to an embodiment of the present invention;

FIG. 3 is a flow chart of generating a first communication behavior waveform graph using first communication behavior feature values in conjunction with time granularity, in accordance with an embodiment of the present invention;

fig. 4 is a flowchart of a method for dividing cell scene categories according to another embodiment of the present invention;

fig. 5 is a communication waveform diagram of a cell of the type of an office building in accordance with an embodiment of the present invention;

FIG. 6 is a table of the structure of a deep learning network model in an embodiment of the present invention;

fig. 7 is a schematic structural diagram of a device for dividing cell scene categories according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of a device for dividing cell scene categories according to another embodiment of the present invention;

fig. 9 is an internal structural view of a computer device in one embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The method for dividing the cell scene categories can be applied to the application environment shown in fig. 1. In the figure, the computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing key quality indicator data of the cell. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement obtaining key performance indicator data for a cell and classifying a scene category of the cell based on the key performance indicator data.

Referring to fig. 2, fig. 2 is a flowchart of a method for dividing a cell scene category according to an embodiment of the present invention, where the method for dividing a cell scene category includes the following steps:

step S210: the method comprises the steps of obtaining first key performance index data of a cell to be divided, and obtaining first communication behavior characteristic data of the cell to be divided from the first key performance index data.

In this step, Key Performance Indicator (KPI) data refers to network management assessment Indicator data of a cell base station, which is used for describing user communication quality parameters; the communication behavior characteristic data comprises a plurality of field data representing communication behaviors in the key performance index data, the correlation among the field data is large, and the field data can comprise the field data such as throughput data, the communication rate, the physical resource block utilization rate and the like of a cell base station. Specifically, the server obtains key performance index data of the cells to be divided within a certain time period, and extracts field data related to the communication behavior from the key performance index data as communication behavior characteristic data.

Step S220: carrying out principal component dimensionality reduction on the relevant data of each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of a cell to be divided;

in this step, because the communication behavior characteristic data itself has stability and the change of the communication behavior characteristic data has a certain rule, the principal component analysis coefficient matrix can be determined according to the data characteristics of the communication behavior characteristic data itself; specifically, the server multiplies first communication behavior characteristic data in first key performance indicator data of the cell to be divided by a preset principal component coefficient matrix, and dimension reduction is carried out on the first communication behavior characteristic data to obtain a corresponding one-dimensional first communication behavior characteristic value.

Step S230: and generating a first communication behavior oscillogram by utilizing the first communication behavior characteristic values and combining the time granularity.

In this step, the server combines the obtained communication behavior characteristic value with the time dimension to generate a communication behavior oscillogram related to time. Specifically, the key performance indicator data may be derived by using time as a dimension, a one-dimensional communication behavior feature value representing communication characteristics of the cell at a certain time is obtained from communication behavior feature data extracted from the key performance indicator data after a principal component dimension reduction operation, and the server may generate a two-dimensional coordinate waveform diagram arranged according to time by using the communication behavior feature value in combination with the time dimension, where the two-dimensional coordinate waveform diagram represents a symbolic communication behavior distribution of the cell.

Step S240: and acquiring the scene category of the cell to be divided according to the first communication behavior oscillogram.

In this step, after the communication behavior oscillogram of the cell to be divided is obtained, the scene category of the cell to be divided can be divided according to the graphic features of the communication behavior oscillogram. Specifically, given the graphic features of the communication behavior oscillograms of different scene categories, the distances from the communication behavior oscillogram of the cell to be divided to the graphic features of different scene categories may be calculated, and the communication behavior oscillogram of the cell to be divided is divided into the scene category with the smallest distance from the corresponding graphic feature.

According to the method for dividing the cell scene categories, the key performance index data of the cell are obtained, the communication behavior characteristic value is obtained through dimensionality reduction of field data related to the communication behavior characteristic in the key performance index data, the communication behavior oscillogram of the cell is further obtained according to the communication behavior characteristic value, the key performance index data of the cell are converted into the visual communication behavior oscillogram, the communication behavior oscillogram is used as the dividing basis of the cell scene categories, individual errors generated by artificial division based on coarse granularities of geographic environment, coverage factors, business characteristics and the like of the cell are effectively avoided, the accurate division of the cell scene categories is achieved, the cell scene categories are accurately divided into the supporting basis of fine adjustment and optimization networks, and planning construction and adjustment of the networks are effectively guided.

Optionally, the communication behavior feature data includes at least one of a network throughput, an RRC (Radio Resource Control) connection indication value, an uplink channel physical Resource block utilization rate, a downlink channel physical Resource block utilization rate, an uplink service information physical Resource block occupancy rate, a downlink service information physical Resource block occupancy rate, and a number of downlink successful transmission initial transport blocks.

Referring to fig. 3, fig. 3 is a flowchart of generating a first communication behavior waveform diagram by using each first communication behavior feature value and combining time granularity in an embodiment of the present invention, and in this embodiment, the step of generating the first communication behavior waveform diagram by using each first communication behavior feature value and combining time granularity includes the following steps:

step S231: and generating a two-dimensional array by utilizing the first communication behavior characteristic value and combining the time granularity.

In this step, the key performance indicator data may be derived by using time as a dimension, after the communication behavior feature data obtained from the key performance indicator data is subjected to principal component dimension reduction operation, a one-dimensional communication behavior feature value is obtained to characterize the communication behavior feature of a cell at a certain time, and the server combines the communication behavior feature values belonging to the same cell and at different times with the corresponding time to generate a two-dimensional array.

Step S232: and generating the original image of the communication behavior by taking the time granularity as an abscissa and the characteristic value of the communication behavior as an ordinate.

In this step, the server generates a two-dimensional coordinate waveform diagram with time as an abscissa and a communication behavior characteristic value as an ordinate according to the two-dimensional data, and the two-dimensional coordinate waveform diagram is an original image of the communication behavior.

Step S231: and removing noise in the original image of the communication behavior, and determining the original image of the communication behavior after the noise is removed as a first communication behavior oscillogram.

In the step, the noise in the original image of the communication behavior comprises interference factors such as a two-dimensional coordinate axis, an image frame, background color and the like; because the existing picture generation tool generally has interference factors such as coordinate axes and picture frames for the generated picture and can cause noise influence on picture identification, the waveform picture of a pure image is obtained by removing the noise in the original image of the communication behavior. Specifically, interference noise such as coordinate axes and frames in the original image of the communication behavior can be eliminated by utilizing Python programming, a waveform picture of a pure image is obtained, accuracy of the waveform picture of the communication behavior for characterizing the communication behavior is improved, and accuracy of the classification of the cell scene classes is improved when a cell scene class classification model is trained by adopting the waveform picture of the communication behavior in the follow-up process.

In the embodiment, the communication behavior oscillogram is generated to change the multidimensional data with complex data into the visual picture data, so that the communication behavior characteristics of different cells are visualized and digitalized, the data complexity is greatly reduced, the problem of inconsistent multidimensional data dimensions is solved, and meanwhile, the later-stage error correction and updating of the cell scene classification model are facilitated.

In one embodiment, before the step of performing principal component dimension reduction on the data related to each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of a cell to be divided, the method further includes the following steps: acquiring second key performance index data of the sample cell, and acquiring second communication behavior characteristic data of the cell to be divided from the second key performance index data; and performing principal component analysis on the second communication behavior characteristic data, and extracting a preset principal component coefficient matrix.

In the network information system, historical key performance index data of a plurality of cell base stations in the whole network can be derived by taking time as a dimension to be transmitted to a server as sample data, the server obtains the key performance index data as the sample data, then carries out bivariate correlation analysis on each field data and a communication behavior field, if the correlation between a certain field data in the key performance index data and the communication behavior field is lower than a preset threshold value, the field data is deleted from the key performance index data, after deleting the field data in the key performance index data, the correlation between the field data and the communication behavior field is lower than the preset threshold value, the remaining field data in the key performance index data is taken as communication behavior characteristic data, and the server carries out principal component analysis to reduce dimension by taking the obtained communication behavior characteristic data of the plurality of cell base stations in the whole network as reference data, and screening the one-dimensional element according to the coefficient of the principal component analysis to obtain a preset principal component coefficient matrix for analyzing the key performance index data of the cell to be divided.

In this embodiment, second key performance indicator data of the sample cell is obtained from the sample cell, principal component analysis and dimension reduction are performed on field data related to communication behavior characteristics in the key performance indicator data, and a one-dimensional element is screened according to coefficients to obtain a preset principal component coefficient matrix. When the class of the cell scene is divided, the first communication behavior characteristic data in the first key performance index data of the cell to be divided is multiplied by the preset principal component coefficient matrix, so that the dimension of the first communication behavior characteristic data can be reduced rapidly to obtain a first communication behavior characteristic value, and the class dividing efficiency of the cell scene is improved.

In one embodiment, the step of obtaining the scene category of the cell to be divided according to the first communication behavior oscillogram includes the following steps: and inputting the first communication behavior oscillogram into a pre-trained deep learning network model, and acquiring the scene category of the cell to be divided, which is output by the pre-trained deep learning network model.

In this embodiment, for the deep learning network model trained in advance, the communication behavior waveform map generated from the key performance indicator data is already set as an input item of the deep learning network model, and the scene category of the cell is set as an output item, so that the first communication behavior waveform map of the cell to be partitioned is input to the deep learning network model trained in advance, and the scene category of the cell to be partitioned is obtained. By applying the deep learning network model, the scene types of the cell can be divided under the condition of acquiring the key performance index data of the cell, so that decision assistance is provided for the optimization and construction of the cell network.

In one embodiment, after the step of performing principal component analysis on each piece of second communication behavior feature data, the method further includes the following steps: acquiring a second communication behavior characteristic value; generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity; screening a communication behavior waveform sample diagram from each second communication behavior waveform diagram, and acquiring a scene type corresponding to each communication behavior waveform sample diagram; and training the deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category.

In this embodiment, after extracting the communication behavior feature data from the historical key performance indicator data of the plurality of cell base stations in the whole network in the network information system, the server extracts the communication behavior feature data from the key performance indicator data as sample data, then performs principal component analysis on the obtained communication behavior feature data of the plurality of cell base stations in the whole network as reference data to perform dimension reduction, after performing dimension reduction on the communication behavior feature data of each multi-dimension, screens out a one-dimensional element according to the coefficient to obtain a preset principal component coefficient matrix and simultaneously obtain a one-dimensional communication behavior feature value corresponding to each communication behavior feature data, the server combines the obtained communication behavior feature value with a time dimension to generate a communication behavior oscillogram related to time, and then screens out a plurality of communication behavior oscillograms with obvious communication behavior distribution characteristics from each communication behavior oscillogram as training sample graphs, and marking corresponding scene category labels on each communication behavior waveform sample graph, inputting the scene category labels as training samples into the deep learning network model for supervised training, and improving the accuracy of the deep learning network model in the classification of the cell scene categories. The step of obtaining the second communication behavior oscillogram is the same as the step of obtaining the first communication behavior oscillogram, a two-dimensional array can be generated by using the second communication behavior characteristic value and combining time granularity, then an original communication behavior image is generated by using the time granularity as an abscissa and the communication behavior characteristic value as an ordinate, noise in the original communication behavior image is removed, and the original communication behavior image after the noise is removed is determined as the second communication behavior oscillogram.

In one embodiment, the step of screening the communication behavior waveform sample diagrams from the second communication behavior waveform diagrams and obtaining the scene types corresponding to the communication behavior waveform sample diagrams includes the following steps: acquiring the number of target scene categories; according to the number of the target scene categories, dividing each second communication behavior oscillogram into different scene clusters with the number being the number of the target scene categories by utilizing a clustering algorithm, and acquiring each scene cluster center and a scene category corresponding to each scene cluster center; taking a second communication behavior waveform diagram with the distance value from each scene clustering center smaller than a preset distance value as a traffic behavior waveform sample diagram; and determining the scene type corresponding to the traffic behavior waveform sample graph according to the scene type corresponding to each scene clustering center.

In this embodiment, the number of target scene categories refers to the number of scene categories; after the server determines the number of the target scene categories, the communication behavior oscillograms with similar characteristics can be divided into one cluster by using a clustering algorithm, so that each communication behavior oscillogram is divided into different scene clusters with the number of the target scene categories respectively, a corresponding scene cluster center is obtained, the scene category of each scene cluster is determined according to the communication behavior characteristics, then a circle region is drawn by taking each scene cluster center as the center and taking a preset distance value as a radius, the communication behavior oscillograms in each circle region are taken as a passing behavior waveform sample map, a plurality of communication behavior oscillograms with obvious communication behavior distribution characteristics are rapidly screened as training samples, the cost of manually screening the training samples is saved, the screening of the training samples is more objective and accurate, and the accuracy of the division of the small-area scene categories is improved.

In one embodiment, the step of obtaining each scene clustering center and the scene category corresponding to each scene clustering center by dividing each second communication behavior oscillogram into different scene clusters with the number of target scene categories by using a clustering algorithm includes the following steps: randomly selecting second communication behavior oscillograms with the number being the number of the target scene categories as a first clustering center of each scene cluster; calculating the distance value from the remaining second communication behavior oscillograms to each first clustering center, dividing each remaining second communication behavior oscillogram into a scene clustering with the smallest distance value from the first clustering center, and obtaining the scene clustering results of the division of each second communication behavior oscillogram into different scene clustering; calculating a second clustering center of each scene cluster according to the scene clustering result, judging whether the first clustering center is equal to the second clustering center, if so, taking each second clustering center as each scene clustering center, and determining the scene type of each scene clustering center; and if the first clustering center is not equal to the second clustering center, taking the second clustering center as the first clustering center, and skipping to execute calculation of the distance value from the remaining second communication behavior oscillogram to each first clustering center.

In this embodiment, K second communication behavior oscillograms are randomly selected from the plurality of communication behavior oscillograms as a first clustering center, where K is the number of categories of the target scene, and then the distance between each second communication behavior oscillogram and the first clustering center is calculated, so that the second communication behavior oscillograms are classified into the cluster where the first clustering center closest to the second communication behavior oscillogram is located. And calculating the average value of the newly formed second communication behavior oscillograms of each cluster to obtain a second cluster center, finishing clustering if the cluster centers of two adjacent clusters are not changed, regarding the second cluster center as the first cluster center if the cluster centers of two adjacent clusters are changed, dividing each communication behavior oscillogram into different cluster center categories again, and iteratively calculating the cluster centers until the cluster centers of two adjacent clusters are not changed.

Referring to fig. 4, fig. 4 is a flowchart of a method for dividing a cell scene category in another embodiment of the present invention, in this embodiment, the method for dividing a cell scene category includes the following steps:

step S301: and acquiring second key performance index data of the sample cell, and acquiring second communication behavior characteristic data of the sample cell from the second key performance index data.

In the step, historical key performance index data of a plurality of cell base stations in the whole network in one year can be derived in a network information system by taking time as a dimension to serve as sample data to a server, after a large amount of sample key performance index data are collected, the server basically cleans the sample key performance index data, zero values in the sample key performance index data are filled with zero, repeated items in the sample key performance index data are deleted, the influence of data noise is reduced to the maximum extent, and meanwhile, field data in the field data are clustered by taking relevant fields of network throughput, call-on rate and the like and communication characteristic behaviors of users in cells as cores, and noise fields are removed.

Specifically, each key performance indicator data includes fields including: time, Cell name CGI (Cell Global Identifier), antenna angle, Cell bandwidth, operating frequency band, network throughput, average throughput of uplink physical resource blocks, average throughput of downlink physical resource blocks, maximum number of RRC connections, number of RRC connection establishment requests, number of RRC connection establishment successes, average utilization of uplink physical resource blocks, average utilization of downlink physical resource blocks, occupancy of uplink service information physical resource blocks, occupancy of downlink service information physical resource blocks, average uplink interference level of physical resource blocks, number of QPSK mode downlink successful initial transmission blocks, number of 16QAM mode downlink successful initial transmission blocks, number of 64QAM mode downlink successful initial transmission blocks, E-RAB drop rate, and wireless drop rate.

After a large amount of sample key performance indicator data are collected, a server respectively carries out bivariate correlation analysis on field data and communication behavior fields, such as network throughput, in the key performance indicator data, field data with correlation lower than a threshold value and without generating large change on cell communication behavior characteristics, such as time, cell name CGI, antenna angle, cell bandwidth and working frequency field, in the key performance indicator data are removed, influence of data noise is reduced, the field data with low correlation with the communication behavior fields are finally screened out to serve as communication behavior characteristic data for research, and eleven field data comprise network throughput, maximum RRC connection number, RRC connection establishment request number, RRC connection establishment success number, data with correlation higher than 0.5 threshold value, The method comprises the following steps of uplink channel physical resource block utilization rate, downlink channel physical resource block utilization rate, uplink service information physical resource block occupancy rate, downlink service information physical resource block occupancy rate, QPSK mode downlink successful transmission initial transmission block quantity, 16QAM mode downlink successful transmission initial transmission block quantity and 64QAM mode downlink successful transmission initial transmission block quantity. And (3) performing principal component analysis by taking communication behavior characteristic data in the key performance index data with hours as dimensionalities of cell base stations of the whole network of a year as reference data to obtain a communication behavior characteristic value.

Step S302: and performing principal component analysis on each second communication behavior characteristic data to obtain a second communication behavior characteristic value and extract a preset principal component coefficient matrix.

In this step, the communication behavior feature data includes a plurality of field data related to the communication behavior; the server performs principal component analysis to reduce dimensions by taking the acquired communication behavior characteristic data of the cell base stations in the whole network as reference data, obtains a one-dimensional communication behavior characteristic value after reducing the dimensions of each multi-dimensional communication behavior characteristic data, and screens out one-dimensional elements according to coefficients of the principal component analysis to obtain a preset principal component coefficient matrix. The preset principal component coefficient matrix is specifically as follows: network throughput 0.943+ maximum number of RRC connections 0.893+ number of RRC connection establishment requests 0.866+ number of RRC connection establishment successes 0.868+ average utilization of upstream PRBs 0.881+ average utilization of downstream PRBs 0.948+ occupancy of upstream traffic information PRBs 0.864+ occupancy of downstream traffic information PRBs 0.955+ 0.947+ initial number of successful transmissions in QPSK mode downlink 0.884+ initial number of successful transmissions in QAM mode downlink TB 0.766.

Step S303: and generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity.

In this step, after bivariate correlation analysis and principal component analysis operations, the key performance indicator data can use one-dimensional communication behavior characteristic values to represent communication characteristics of a cell at a certain time, the server combines the communication behavior characteristic values belonging to the same cell at different times within the same day with corresponding time to generate two-dimensional arrays arranged according to a 24-hour sequence, generates and forms a 24 x 24 two-dimensional coordinate waveform diagram with time as a horizontal coordinate and the communication behavior characteristic values as a vertical coordinate according to the two dimensions, removes interference noise such as two-dimensional coordinate axes, image borders, background colors and the like in the two-dimensional coordinate waveform diagram, and obtains a pure image waveform diagram, thereby obtaining the communication behavior waveform diagram representing the communication behavior characteristics of all cells of the whole network at every day throughout the year.

Step S304: and acquiring the number of target scene categories.

In this step, the number of the traditional scene categories is usually from 6 to 13, on the basis, the clustering algorithm can be used to divide each communication behavior oscillogram into multiple categories of scene categories according to the number of the scene categories from 6 to 13, to obtain clustering results from 6 to 13, to compare indexes such as the intensity of the communication behavior oscillograms in each scene category and the difference between various scene categories in each clustering result, and finally, to determine the number of the scene categories with the best overall quality of the clustering results as 9 categories, and to determine the number of the target scene categories as 9 categories.

Step S305: and according to the number of the target scene categories, dividing each second communication behavior oscillogram into different scene clusters with the number of the target scene categories by utilizing a clustering algorithm, and acquiring each scene cluster center and the scene category corresponding to each scene cluster center.

In this step, the number of the target scene categories is determined to be 9 categories, the second communication behavior oscillogram is divided into 9 scene clusters, the scene category corresponding to each scene cluster is determined according to the characteristics of the communication behavior oscillogram of each scene cluster, and finally, the 9 scene categories are defined as follows: wave type, fault type, shift area office building type, discrete type, business district type, bar type, accommodation district type, and village type. Referring to fig. 5, fig. 5 is a communication waveform diagram of an office building-type cell according to an embodiment of the present invention, in the diagram, communication behaviors between 0 point and 8 point are less, a communication behavior between 8 point and 19 point starts to rise and then reaches a peak value of a duration time, and a communication behavior between 19 point and 0 point is in a greatly decreasing trend, so as to satisfy communication behavior characteristics of the office building cell.

Step S306: and taking the second communication behavior waveform diagram with the distance value from each scene clustering center smaller than the preset distance value as a traffic behavior waveform sample diagram.

Step S307: and determining the scene type corresponding to the traffic behavior waveform sample graph according to the scene type corresponding to each scene clustering center.

Step S308: and training the deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category.

In this step, the server may use the communication behavior waveform sample map and the scene category corresponding to the communication behavior waveform sample map as an input item of the deep learning network model, and the deep learning network model performs analysis training according to the input communication behavior waveform sample map, outputs an identification result and compares the identification result with the scene category corresponding to the communication behavior waveform sample map, and reversely corrects parameters of each layer of network in the deep learning network model according to the comparison result, where a structure table of the deep learning network model is shown in fig. 6, a pooling layer in the deep learning network model uses two types, namely a maximum pooling layer and a random pooling layer, a step size of the random pooling layer is set to a root number 2, the number of iterations of training of the deep learning network model is set to 100000 times, an Adadelta optimizer uses an Adadelta optimizer, a learning rate is set to 0.001, and a regular item coefficient is set to 0.0005. By using the random pooling layer, the fuzzy recognition effect of the deep learning network model is improved, and the robustness of the deep learning network model is improved.

Step S309: the method comprises the steps of obtaining first key performance index data of a cell to be divided, and obtaining first communication behavior characteristic data of the cell to be divided from the first key performance index data.

In this step, the server performs bivariate correlation analysis on the sample data, and determines that the communication behavior characteristic data in the key performance indicator data includes network throughput, maximum number of RRC connections, number of times of RRC connection establishment requests, number of times of RRC connection establishment success, utilization rate of uplink physical resource blocks, utilization rate of downlink physical resource blocks, occupancy rate of uplink traffic information physical resource blocks, occupancy rate of downlink traffic information physical resource blocks, number of QPSK mode downlink successful transmission initial transmission blocks, number of 16QAM mode downlink successful transmission initial transmission blocks, and number of 64QAM mode downlink successful transmission initial transmission blocks, so that the field data can be directly acquired from the first key performance indicator data of the cell to be divided as the first communication behavior characteristic data.

Step S310: and performing principal component dimensionality reduction on the relevant data of each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of the cell to be divided.

In the step, the server multiplies first communication behavior characteristic data in first key performance index data of the cell to be divided by a preset principal component coefficient matrix, and dimension reduction is carried out on the first communication behavior characteristic data to obtain a corresponding one-dimensional first communication behavior characteristic value.

Step S311: and generating a first communication behavior oscillogram by utilizing the first communication behavior characteristic values and combining the time granularity.

Step S312: and inputting the first communication behavior oscillogram into a pre-trained deep learning network model, and acquiring the scene category of the cell to be divided, which is output by the pre-trained deep learning network model.

In the embodiment, by acquiring key performance indicator data serving as sample data, a communication behavior characteristic value is acquired by dimensionality reduction of field data related to communication behavior characteristics in the sample data, a communication behavior oscillogram of a cell is further acquired, a training sample is screened from the communication behavior oscillogram to train a deep learning network model, key performance indicator data of the cell to be divided is subsequently acquired, communication behavior characteristic data can be acquired from the key performance indicator data of the cell to be divided, and the communication behavior oscillogram is generated from the communication behavior characteristic data, so that the cell is quickly divided into different scene categories according to the communication behavior oscillogram, the synchronous updating efficiency of the scene categories of the cell is improved, and an important basis is provided for customizing a construction scheme and determining a network optimization strategy.

In one embodiment, after the step of obtaining the scene category of the to-be-divided cell output by the cell scene category division model, the network state parameter of the to-be-divided cell may be further adjusted according to the scene category of the to-be-divided cell to optimize the allocation of the cell base station, and the step of adjusting the network state parameter of the to-be-divided cell according to the scene category of the to-be-divided cell specifically includes the following steps: determining a corresponding characteristic service set according to the scene category of the target cell; acquiring quality parameters and reference values of service perception factors of all feature services in a feature service set; determining a perception quality influence factor from the business perception factors according to the quality parameters of the business perception factors and the reference value, and matching target network parameters influencing the perception quality influence factor; and adjusting the target network parameters according to the preset expected performance index of the target cell.

In the embodiment, the situation of key performance index data of each cell base station every day is monitored in real time, the cells with changed scene categories are divided in real time, meanwhile, the feature service set under the scene categories is obtained according to the division result of the scene categories, the perception quality influence factor influencing the use of the feature service experience of a terminal user is determined, the network parameter influencing the perception quality influence factor is adjusted by combining the expected performance index of the cell to be divided, the perception quality influence factor aiming at the feature service under the scene categories of the cell is realized, the key network parameter is accurately adjusted, the service performance of the cell network is effectively improved, the network optimization effect is improved, and the fine adjustment and optimization of the network are realized.

In one embodiment, feature services in the same scene category are the same, and feature services in different scene categories are different, so that a feature service set corresponding to each scene category can be obtained in advance, and the specific steps include: acquiring the access volume and the flow of various services in the full service under the scene type of a target cell; respectively calculating the weight values of various services in the full-volume service according to the access volume and the flow of various services; and determining the service with the weight value ratio exceeding a preset threshold value as the corresponding characteristic service under the scene category of the target cell, and acquiring a characteristic service set.

Specifically, access volume and flow of each type of service in the full-volume service under each type of scene category are counted, based on the difference between the access volume and the flow of each type, the weight value of each type of service under each type of scene category is obtained, and according to the size of the weight value, the service with the weight value ratio exceeding a certain threshold is selected as the feature service of the scene, so that a feature service set is formed. By determining the feature service set of each scene category, when the scene category classification result of the cell is obtained, the feature service set of the cell can be rapidly determined, so that the perception quality influence factor and the corresponding network parameter under the scene category are obtained, and the optimization efficiency of the system is improved.

It should be understood that, although the steps in the flowcharts of fig. 2 to 4 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-4 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

According to the method for dividing the cell scene categories, the invention further provides a device for dividing the cell scene categories, and embodiments of the device for dividing the cell scene categories according to the invention are described in detail below.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a device for dividing a cell scene category according to an embodiment of the present invention, in this embodiment, the device for dividing a cell scene category includes:

the communication behavior characteristic data acquiring module 410 is configured to acquire first key performance indicator data of a cell to be divided, and acquire the first communication behavior characteristic data of the cell to be divided from the first key performance indicator data;

a communication behavior characteristic value obtaining module 420, configured to perform principal component dimension reduction on the data related to each first communication behavior by using a preset principal component coefficient matrix, so as to obtain a first communication behavior characteristic value of a cell to be partitioned;

a communication behavior waveform diagram obtaining module 430, configured to generate a first communication behavior waveform diagram by using each first communication behavior feature value and combining time granularity;

and a scene category dividing module 440, configured to obtain a scene category of the cell to be divided according to the first communication behavior oscillogram.

In the dividing device for the cell scene categories, the key performance index data of the cell is obtained, the communication behavior characteristic value is obtained through dimensionality reduction of the field data related to the communication behavior characteristic in the key performance index data, the communication behavior oscillogram of the cell is further obtained according to the communication behavior characteristic value, and the communication behavior oscillogram is used as the dividing basis of the scene categories of the cell, so that the accuracy of the division of the cell scene categories is greatly improved, and an important basis is provided for customizing a planning construction scheme and determining a network optimization strategy.

In one embodiment, the communication behavior oscillogram obtaining module 430 is configured to generate a two-dimensional array by using the first communication behavior feature value in combination with the time granularity; generating a communication behavior original image by taking the time granularity as an abscissa and the communication behavior characteristic value as an ordinate; and removing noise in the original image of the communication behavior, and determining the original image of the communication behavior after the noise is removed as a first communication behavior oscillogram.

In one embodiment, the scene classification module 440 is configured to input the first communication behavior oscillogram into a pre-trained deep learning network model, and obtain a scene classification of a cell to be classified output by the pre-trained deep learning network model.

Referring to fig. 8, fig. 8 is a schematic structural diagram of a device for dividing a cell scene category according to another embodiment of the present invention, in this embodiment, the device for dividing a cell scene category further includes a principal component coefficient matrix obtaining module 450, configured to obtain second key performance indicator data of a sample cell, and obtain second communication behavior feature data of the sample cell from the second key performance indicator data; and performing principal component analysis on the second communication behavior characteristic data, and extracting a preset principal component coefficient matrix.

In one embodiment, the apparatus for classifying cell scene categories further includes a model training module 460, configured to obtain a second communication behavior feature value; generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity; screening a communication behavior waveform sample diagram from each second communication behavior waveform diagram, and acquiring a scene type corresponding to each communication behavior waveform sample diagram; and training the deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category.

In one embodiment, the model training module 460 obtains the number of target scene categories; according to the number of the target scene categories, dividing each second communication behavior oscillogram into different scene clusters with the number being the number of the target scene categories by utilizing a clustering algorithm, and acquiring each scene cluster center and a scene category corresponding to each scene cluster center; taking a second communication behavior waveform diagram with the distance value from each scene clustering center smaller than a preset distance value as a traffic behavior waveform sample diagram; and determining the scene type corresponding to the traffic behavior waveform sample graph according to the scene type corresponding to each scene clustering center.

In one embodiment, the communication behavior characteristic data includes at least one of network throughput, RRC connection indication value, uplink channel physical resource block utilization rate, downlink channel physical resource block utilization rate, uplink traffic information physical resource block occupancy rate, downlink traffic information physical resource block occupancy rate, and downlink successful transmission initial transport block number.

For specific limitation of the device for dividing the cell scene categories, reference may be made to the above limitation on the method for dividing the cell scene categories, and details are not described herein again. All or part of the modules in the division device of the cell scene category can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing key quality indicator data of the cell. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of partitioning cell context classes.

Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, the present invention further provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the following steps when executing the computer program:

carrying out principal component dimensionality reduction on the relevant data of each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of a cell to be divided;

generating a first communication behavior oscillogram by utilizing the first communication behavior characteristic values and combining time granularity;

In one embodiment, when the processor executes the computer program to implement the step of generating the first communication behavior oscillogram by using each first communication behavior feature value and combining the time granularity, the following steps are specifically implemented:

generating a two-dimensional array by utilizing the first communication behavior characteristic value and combining time granularity;

generating a communication behavior original image by taking the time granularity as an abscissa and the communication behavior characteristic value as an ordinate;

and removing noise in the original image of the communication behavior, and determining the original image of the communication behavior after the noise is removed as a first communication behavior oscillogram.

In one embodiment, when the processor executes a computer program to implement the step of performing principal component dimension reduction on the data related to each first communication behavior by using a preset principal component coefficient matrix to obtain a first communication behavior characteristic value of a cell to be partitioned, the following steps are specifically implemented: acquiring second key performance indicator data of the sample cell, and acquiring second communication behavior characteristic data of the sample cell from the second key performance indicator data; and performing principal component analysis on the second communication behavior characteristic data, and extracting a preset principal component coefficient matrix.

In one embodiment, when the processor executes the computer program to implement the step of obtaining the scene category of the cell to be divided according to the first communication behavior oscillogram, the following steps are specifically implemented: and inputting the first communication behavior oscillogram into a pre-trained deep learning network model, and acquiring the scene category of the cell to be divided, which is output by the pre-trained deep learning network model.

In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring a second communication behavior characteristic value; generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity; screening a communication behavior waveform sample diagram from each second communication behavior waveform diagram, and acquiring a scene type corresponding to each communication behavior waveform sample diagram; and training the deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category.

In one embodiment, when the processor executes the computer program to realize the step of screening the communication behavior waveform sample diagrams from the second communication behavior waveform diagrams and acquiring the scene types corresponding to the communication behavior waveform sample diagrams, the following steps are specifically realized: acquiring the number of target scene categories; according to the number of the target scene categories, dividing each second communication behavior oscillogram into different scene clusters with the number being the number of the target scene categories by utilizing a clustering algorithm, and acquiring each scene cluster center and a scene category corresponding to each scene cluster center; taking a second communication behavior waveform diagram with the distance value from each scene clustering center smaller than a preset distance value as a traffic behavior waveform sample diagram; and determining the scene type corresponding to the traffic behavior waveform sample graph according to the scene type corresponding to each scene clustering center.

In one embodiment, the present invention further provides a computer readable storage medium having a computer program stored thereon, the computer program when executed by a processor implementing the steps of:

In one embodiment, when the computer program is executed by the processor to implement the step of generating the first communication behavior oscillogram by using the first communication behavior feature values and combining the time granularity, the following steps are specifically implemented: generating a two-dimensional array by utilizing the first communication behavior characteristic value and combining time granularity; generating a communication behavior original image by taking the time granularity as an abscissa and the communication behavior characteristic value as an ordinate; and removing noise in the original image of the communication behavior, and determining the original image of the communication behavior after the noise is removed as a first communication behavior oscillogram.

In one embodiment, when the computer program is executed by the processor to implement the step of performing principal component dimension reduction on the data related to each first communication behavior by using the preset principal component coefficient matrix to obtain the first communication behavior characteristic value of the cell to be divided, the following steps are specifically implemented: acquiring second key performance indicator data of the sample cell, and acquiring second communication behavior characteristic data of the sample cell from the second key performance indicator data; and performing principal component analysis on the second communication behavior characteristic data, and extracting a preset principal component coefficient matrix.

In one embodiment, when the computer program is executed by the processor to implement the step of obtaining the scene category of the cell to be divided according to the first communication behavior oscillogram, the following steps are specifically implemented: and inputting the first communication behavior oscillogram into a pre-trained deep learning network model, and acquiring the scene category of the cell to be divided, which is output by the pre-trained deep learning network model.

In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a second communication behavior characteristic value; generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity; screening a communication behavior waveform sample diagram from each second communication behavior waveform diagram, and acquiring a scene type corresponding to each communication behavior waveform sample diagram; and training the deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category.

In one embodiment, when the computer program is executed by the processor to implement the steps of screening the communication behavior waveform sample diagrams from the second communication behavior waveform diagrams and obtaining the scene types corresponding to the communication behavior waveform sample diagrams, the following steps are specifically implemented: acquiring the number of target scene categories; according to the number of the target scene categories, dividing each second communication behavior oscillogram into different scene clusters with the number being the number of the target scene categories by utilizing a clustering algorithm, and acquiring each scene cluster center and a scene category corresponding to each scene cluster center; taking a second communication behavior waveform diagram with the distance value from each scene clustering center smaller than a preset distance value as a traffic behavior waveform sample diagram; and determining the scene type corresponding to the traffic behavior waveform sample graph according to the scene type corresponding to each scene clustering center.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method for dividing cell scene categories is characterized by comprising the following steps:

acquiring first key performance index data of a cell to be divided, wherein the first key performance index data is derived by a time dimension, and acquiring first communication behavior characteristic data of the cell to be divided from the first key performance index data;

acquiring second key performance indicator data derived by a time dimension of a sample cell, and acquiring second communication behavior characteristic data of the sample cell from the second key performance indicator data;

performing principal component analysis on each second communication behavior characteristic data, and extracting a preset principal component coefficient matrix;

generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining the time granularity;

acquiring the number of target scene categories;

according to the number of the target scene categories, the second communication behavior oscillogram is divided into different scene clusters with the number being the number of the target scene categories by utilizing a clustering algorithm, and each scene cluster center and a scene category corresponding to each scene cluster center are obtained;

taking the second communication behavior oscillogram with the distance value from each scene clustering center smaller than a preset distance value as a communication behavior oscillogram sample graph;

determining scene types corresponding to the communication behavior waveform sample graph according to the scene types corresponding to the scene clustering centers;

training a deep learning network model by utilizing each communication behavior waveform sample diagram and the corresponding scene category;

inputting the first communication behavior oscillogram into the pre-trained deep learning network model, and acquiring the scene category of the cell to be divided, which is output by the pre-trained deep learning network model.

2. The method for classifying cell scene categories according to claim 1, wherein the step of generating a first communication behavior oscillogram by using each of the first communication behavior feature values and combining time granularity includes the steps of:

3. The method for classifying cell scene categories according to claim 1, wherein after the step of performing principal component analysis on each of the second communication behavior feature data, the method further comprises the following steps:

acquiring a second communication behavior characteristic value;

generating a second communication behavior oscillogram by utilizing the second communication behavior characteristic value and combining time granularity;

and screening a communication behavior waveform sample graph from each second communication behavior waveform graph, and acquiring a scene type corresponding to each communication behavior waveform sample graph.

4. The method for dividing cell scene categories according to any one of claims 1 to 3, wherein the communication behavior characteristic data includes at least one of network throughput, RRC connection indication value, uplink channel physical resource block utilization ratio, downlink channel physical resource block utilization ratio, uplink traffic information physical resource block occupancy, downlink traffic information physical resource block occupancy, and downlink successful transmission initial transport block number.

5. The method for dividing the scene categories of the cell according to claim 1, wherein the step of dividing the second communication behavior oscillogram into different scene clusters with a number of target scene categories by using a clustering algorithm to obtain each scene cluster center and the scene category corresponding to each scene cluster center comprises the following steps:

randomly selecting the second communication behavior oscillograms with the quantity as the number of the target scene categories as a first clustering center of each scene cluster;

calculating the distance value from the remaining second communication behavior oscillograms to each first clustering center, and respectively partitioning each remaining second communication behavior oscillogram into a scene clustering with the smallest distance value from the first clustering center to obtain the scene clustering results of the second communication behavior oscillograms partitioned into different scene clustering:

calculating a second clustering center of each scene cluster according to a scene clustering result, judging whether the first clustering center is equal to the second clustering center, if so, taking each second clustering center as each scene clustering center, and determining a scene type corresponding to each scene clustering center;

and if the first clustering center is not equal to the second clustering center, taking the second clustering center as the first clustering center, and skipping to execute calculation of the distance value from the remaining second communication behavior oscillogram to each first clustering center.

6. The method for dividing the scene categories of the cells according to claim 1, wherein after the step of obtaining the scene categories of the cells to be divided output by the pre-trained deep learning network model, the method further comprises the following steps:

determining a corresponding characteristic service set according to the scene category of the target cell: acquiring quality parameters and reference values of service perception factors of each feature service in the feature service set;

determining a perception quality influence factor from the service perception factors according to the quality parameters of the service perception factors and a reference value, and matching target network parameters influencing the perception quality influence factor;

and adjusting the target network parameters according to the preset expected performance index of the target cell.

7. An apparatus for classifying cell scene categories, comprising:

the communication behavior characteristic data acquisition module is used for acquiring first key performance index data of a cell to be divided, wherein the first key performance index data is derived by time dimension, and acquiring first communication behavior characteristic data of the cell to be divided from the first key performance index data;

the scene category dividing module is used for acquiring second key performance indicator data derived by time dimension of a sample cell and acquiring second communication behavior characteristic data of the sample cell from the second key performance indicator data;

acquiring the number of target scene categories;

8. The apparatus for classifying cell scene categories according to claim 7, wherein the communication behavior oscillogram obtaining module is configured to generate a two-dimensional array by using the first communication behavior feature value and combining the time granularity;

removing noise in the original image of the communication behavior, and determining the original image of the communication behavior after the noise is removed as the first communication behavior oscillogram.

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor when executing the computer program implements the steps of the method for partitioning cell scene classes according to any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for partitioning cell scene classes according to any one of claims 1 to 7.