CN109495318B - Mobile communication network flow prediction method, device and readable storage medium - Google Patents
Mobile communication network flow prediction method, device and readable storage medium Download PDFInfo
- Publication number
- CN109495318B CN109495318B CN201811543654.3A CN201811543654A CN109495318B CN 109495318 B CN109495318 B CN 109495318B CN 201811543654 A CN201811543654 A CN 201811543654A CN 109495318 B CN109495318 B CN 109495318B
- Authority
- CN
- China
- Prior art keywords
- date
- data
- prediction model
- holiday
- flow
- 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.)
- Active
Links
Images
Classifications
-
- 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
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
Abstract
The invention discloses a method, a device and a readable storage medium for predicting mobile communication network flow, which comprise the following steps: acquiring filling data by adopting a preset filling data prediction model according to the collected user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number; obtaining a feature set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data; training a pre-established machine learning model according to the feature set to obtain a flow prediction model; the date to be predicted is input into the flow prediction model to obtain a flow prediction value, so that the problem of single modeling dimension in the prior art can be effectively solved, and the prediction accuracy and the application range of the model can be effectively improved.
Description
Technical Field
The present invention relates to the field of communication technologies and machine learning technologies, and in particular, to a method and an apparatus for predicting mobile communication network traffic, and a readable storage medium.
Background
With the rapid development of mobile communication networks, the current network size becomes increasingly large. The rapidly developed information technology makes the network flow behavior become more and more complex, and video, audio, image data and other network application services transmitted through the network are continuously increased, so that more requirements are provided for the service quality, the network safety and other related performances in the network service process, and how to analyze and predict the change rule of the service flow in the network becomes an important direction for the future network research. Through the network flow prediction, a basis can be provided for network management, planning and maintenance.
The existing scheme only reflects the characteristics of periodicity, historical trend and the like of the flow by using the time series data of the flow, and when the flow is influenced by external factors and periodically changes, the model prediction error is large.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting mobile communication network flow and a readable storage medium, which can effectively solve the problem of single modeling dimension in the prior art and effectively improve the prediction accuracy and the application range of a model.
An embodiment of the present invention provides a method for predicting mobile communication network traffic, including:
acquiring filling data by adopting a preset filling data prediction model according to the collected user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number;
obtaining a feature set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data;
training a pre-established machine learning model according to the feature set to obtain a flow prediction model;
and inputting the date to be predicted into the flow prediction model to obtain a flow prediction value.
As an improvement of the above solution, the method further includes a dividing step of the user data of the preset date:
dividing the collected user data of the preset date into a holiday sequence and a natural day sequence;
dividing the holiday sequence into a holiday flow time sequence and a holiday user number time sequence;
and dividing the natural day sequence into a natural day traffic time sequence and a natural day user number time sequence.
As an improvement of the above solution, the method further includes a step of constructing the padding data prediction model:
the filling data prediction model comprises a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model;
training a pre-established machine learning algorithm according to the holiday flow time sequence to obtain a holiday flow prediction model;
training the machine learning algorithm according to the holiday user number time sequence to obtain a holiday user number prediction model;
training the machine learning algorithm according to the natural daily traffic time sequence to obtain a natural daily traffic prediction model;
and training the machine learning algorithm according to the natural daily user number time sequence to obtain a natural daily user number prediction model.
As an improvement of the above scheme, the acquiring of the filling data by using a preset filling data prediction model according to the collected user data on the preset date specifically includes:
judging the type of the date to be predicted by adopting a crawler system;
determining a date threshold to be filled according to the type of the date to be predicted;
and inputting the threshold value of the date to be filled into a filling data prediction model corresponding to the type of the date to be predicted for prediction to obtain filling data.
Further, the inputting the threshold of the date to be filled into the filling data prediction model corresponding to the type of the date to be predicted to perform prediction to obtain filling data specifically includes:
when the date to be predicted is judged to be the holiday date, taking the date set of all holidays in the time threshold value between the preset date and the date to be predicted as the date threshold value to be filled;
and inputting the threshold value of the date to be filled into the holiday flow prediction model and the holiday user number prediction model to obtain filling data.
As an improvement of the above scheme, the obtaining a feature set according to the padding data and the external data specifically includes:
calculating the related characteristic values of the flow value and the user quantity according to the filling data;
crawling the type of the date to be predicted and weather data as external data according to a crawler system;
and generating a feature set according to the relevant feature value and the external data.
As an improvement of the above scheme, the training of the pre-established machine learning model according to the feature set to obtain the traffic prediction model specifically includes:
taking the user data of the preset date as a training sample;
inputting the feature set into a pre-established machine learning model, and training the machine learning model according to the training sample;
and performing parameter optimization on the machine learning model obtained by training by adopting the training sample to obtain a flow prediction model.
Another embodiment of the present invention correspondingly provides a device for predicting mobile communication network traffic, including:
the data acquisition module is used for acquiring the filling data by adopting a preset filling data prediction model according to the acquired user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number;
the characteristic acquisition module is used for obtaining a characteristic set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data;
the model construction module is used for training a pre-established machine learning model according to the feature set to obtain a flow prediction model;
and the flow prediction module is used for inputting the date to be predicted into the flow prediction model to obtain a flow prediction value.
Compared with the prior art, the mobile communication network traffic prediction method disclosed by the embodiment of the invention acquires the filling data by adopting the preset filling data prediction model according to the acquired user data of the preset date, wherein the filling data is the user data of a time threshold value between the preset date and the date to be predicted, the user data is the time sequence data of a traffic value and a user number, and the feature set is obtained according to the filling data and the external data, wherein the external data comprises the type of the date to be predicted and the weather data, the pre-established machine learning model is trained according to the feature set to obtain the traffic prediction model, the date to be predicted is input into the traffic prediction model to obtain a traffic prediction value, the traffic prediction model is established based on the multi-dimensional data of the type, the traffic, the user number, the weather and the like of the date to be predicted, by increasing more factors influencing flow change in the real environment and particularly considering that the flow forms of holidays and natural days have larger difference, the influence of the holiday factors on the flow change can be effectively eliminated, the robustness of the model can be effectively improved, the problems that the dimension of the model is single and the application range is limited due to the fact that the model is modeled only by adopting the time series data of the flow in the prior art can be effectively solved, and the prediction accuracy and the application range of the model are improved.
Another embodiment of the present invention provides a mobile communication network traffic prediction apparatus, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the mobile communication network traffic prediction apparatus implements the mobile communication network traffic prediction method according to the above embodiment of the present invention.
Another embodiment of the present invention provides a storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute the method for predicting mobile communication network traffic according to the above embodiment of the present invention.
Drawings
Fig. 1 is a flowchart illustrating a method for predicting traffic of a mobile communication network according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a processing flow of padding data according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a traffic prediction apparatus of a mobile communication network according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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.
Referring to fig. 1, which is a flowchart illustrating a method for predicting traffic of a mobile communication network according to an embodiment of the present invention, the method includes:
s11, acquiring filling data by adopting a preset filling data prediction model according to the collected user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number.
Specifically, a crawler system is adopted to judge the type of the date to be predicted;
determining a date threshold to be filled according to the type of the date to be predicted;
and inputting the threshold value of the date to be filled into a filling data prediction model corresponding to the type of the date to be predicted for prediction to obtain filling data.
It should be noted that the preset date threshold is a date recorded in a certain time period according to a chronological order. Wherein, the statistical date threshold value can be divided into holidays and natural days. Holidays are defined as legal holidays and comprise New year, spring festival, Qingming festival, labor festival, early afternoon festival, mid-autumn festival and national celebration festival; the natural day is defined as a normal working day and a double holiday after the holiday removal date.
Referring to fig. 2, which is a schematic view of a processing flow of filling data provided in an embodiment of the present invention, since part of holidays in the legal holiday are lunar calendar holidays, the holiday time changes every year and is combined with a calendar module of a crawler system to obtain the type of the date to be predicted, that is, a holiday or a natural day. And acquiring filling data with a date threshold to be filled as a vacant interval between the date to be predicted and the preset date based on a pre-established filling data prediction model. The filling data prediction model comprises a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model, and the filling data comprises holiday or natural day filling day granularity flow and user number. For example, when it is detected that the date to be predicted is a holiday, the date intervals from the last day of the preset date threshold to the day before the date to be predicted are screened. And screening the holiday dates in the date interval by combining the crawler system and national legal holiday regulations, and arranging all the selected holiday dates according to a time sequence to form a date threshold to be filled. And then, inputting the date threshold value to be filled into a corresponding filling data prediction model for prediction to obtain day granularity flow and the number of users in the day.
In an optional embodiment, when the date to be predicted is judged to be a holiday, taking a date set of all holidays in a time threshold from the preset date to the date to be predicted as the date threshold to be filled;
and inputting the threshold value of the date to be filled into the holiday flow prediction model and the holiday user number prediction model to obtain filling data.
For example, the date to be predicted is 2018.10.7, the last day of the preset date is 2018.9.28 (natural day), and if 2018.10.7 is holiday in combination with the calendar module, the date to be filled threshold is 2018.10.1-2018.10.6 (holiday). And 2018.9.29-2018.9.30 are natural days and do not need to be used as feature input models. Inputting the date threshold value to be filled into a holiday flow prediction model to obtain a flow value of the date to be predicted, and inputting the date threshold value to be filled into a holiday user number prediction model to obtain the number of users of the date to be predicted, namely holiday filling data.
In another optional embodiment, when the date to be predicted is judged to be a natural day, taking a date set of all natural days in a time threshold value between the preset date and the date to be predicted as the date to be filled threshold value;
and inputting the date threshold value to be filled into the natural daily flow prediction model and the natural daily user number prediction model to obtain filling data.
Similarly, the holiday date is removed from the threshold value of the date to be filled of the natural day, and the method is similar to the holiday filling method. For example, the date to be predicted is 2018.10.30, the last day of the historical statistical date is 2018.9.30, 2018.10.30 is a natural day according to the calendar module, holidays in 2018.10.1-2018.10.29 are removed, and the remaining dates are thresholds 2018.10.8-2018.10.29 to be filled. And inputting the date threshold value to be filled into a natural day flow prediction model to obtain a flow value of the date to be predicted, and inputting the date threshold value to be filled into a natural day user number prediction model to obtain the number of users of the date to be predicted, namely natural day filling data.
S12, obtaining a feature set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data. .
Specifically, according to the filling data, calculating a related characteristic value of a flow value and a user number;
crawling the type of the date to be predicted and weather data as external data according to a crawler system;
and generating a feature set according to the relevant feature value and the external data.
The external data are actual environmental factors which are crawled by a crawler system and affect the change of the flow, for example, the type of the date to be predicted is acquired as holidays or natural days by crawling a calendar module, and weather data are acquired by crawling a weather module and comprise weather types and average temperatures, for example, the weather types are sunny days, cloudy days, snowing and raining. The feature set is used as the features of the training model, and the specific feature set content is as follows:
| serial number | Feature(s) |
| 1 | Whether to save holidays |
| 2 | Weather (weather) |
| 3 | Mean temperature |
| 4 | Flow 1 day before the date to be predicted |
| 5 | Flow 2 days before the date to be predicted |
| 6 | Flow 3 days before the date to be predicted |
| 7 | Flow 4 days before the date to be predicted |
| 8 | Flow 5 days before the date to be predicted |
| 9 | Flow 6 days before the date to be predicted |
| 10 | Flow 7 days before the date to be predicted |
| 11 | Flow 8 days before the date to be predicted |
| 12 | 3 balance flow at the previous date to be predicted |
| 13 | 5 balance flow at the previous date to be predicted |
| 14 | Balance flow of 7 scales before date to be predicted |
| 15 | Average flow of 30 balances before date to be predicted |
| 16 | Number of users 1 day before the date to be predicted |
| 17 | Number of users 2 days before the date to be predicted |
| 18 | Number of users 3 days before the date to be predicted |
| 19 | Date to be predictedNumber of users in the first 4 days |
| 20 | Number of users 5 days before the date to be predicted |
| 21 | Number of users 6 days before the date to be predicted |
| 22 | Number of users 7 days before the date to be predicted |
| 23 | Number of users 8 days before the date to be predicted |
| 24 | Average flow of people 1 day before the date to be predicted |
| 25 | Average flow 2 days before the date to be predicted |
| 26 | Average flow 3 days before the date to be predicted |
| 27 | Average flow 4 days before the date to be predicted |
| 28 | Average flow of people 5 days before the date to be predicted |
| 29 | Average flow of people 6 days before the date to be predicted |
| 30 | Average flow of people 7 days before the date to be predicted |
And S13, training a pre-established machine learning model according to the feature set to obtain a flow prediction model.
And S14, inputting the date to be predicted into the flow prediction model to obtain a flow prediction value.
The embodiment of the invention discloses a mobile communication network flow prediction method, which comprises the steps of acquiring filling data by adopting a preset filling data prediction model according to acquired user data of a preset date, wherein the filling data is user data of a time threshold value between the preset date and a date to be predicted, the user data is time sequence data of a flow value and a user number, and a feature set is obtained according to the filling data and external data, wherein the external data comprises the type of the date to be predicted and weather data, a pre-established machine learning model is trained according to the feature set to obtain a flow prediction model, the date to be predicted is input into the flow prediction model to obtain a flow prediction value, and the flow prediction model is established based on multi-dimensional data such as the type, the flow, the user number, the weather and the like of the date to be predicted, by increasing more factors influencing flow change in the real environment and particularly considering that the flow forms of holidays and natural days have larger difference, the influence of the holiday factors on the flow change can be effectively eliminated, the robustness of the model can be effectively improved, the problems that the dimension of the model is single and the application range is limited due to the fact that the model is modeled only by adopting the time series data of the flow in the prior art can be effectively solved, and the prediction accuracy and the application range of the model are improved.
On the basis of the above embodiment, step S11 further includes a dividing step of the user data of the preset date:
dividing the collected user data of the preset date into a holiday sequence and a natural day sequence;
dividing the holiday sequence into a holiday flow time sequence and a holiday user number time sequence;
and dividing the natural day sequence into a natural day traffic time sequence and a natural day user number time sequence.
It can be understood that the total network flow and the number of users on a preset date are collected, the data of the holiday and the holiday are extracted to generate the time sequence data of the holiday and the holiday total network flow and the user granularity of several days, and the remaining date generates the natural day total network flow and the user granularity of several days. The holiday sequence is flow data and user number data which are recorded in a certain time period according to the holiday time sequence, and is divided into a holiday flow time sequence and a holiday user number time sequence. The natural day sequence is flow data and user number data which are recorded according to the natural day time sequence in a certain time period and is divided into a natural day flow time sequence and a natural day user number time sequence. The user data of the preset date can be understood as historical statistical data, and the filling data can be understood as recent statistical data.
Further, the step of constructing the padding data prediction model in step S11 is:
the filling data prediction model comprises a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model;
training a pre-established machine learning algorithm according to the holiday flow time sequence to obtain a holiday flow prediction model;
training the machine learning algorithm according to the holiday user number time sequence to obtain a holiday user number prediction model;
training the machine learning algorithm according to the natural daily traffic time sequence to obtain a natural daily traffic prediction model;
and training the machine learning algorithm according to the natural daily user number time sequence to obtain a natural daily user number prediction model.
Preferably, the machine learning model may be a Holt-Winters cubic exponential smoothing algorithm.
In the embodiment, a method for respectively modeling the natural day and the holiday is adopted, the flow forms of the holiday and the natural day are considered to have large difference, the time of a part of holidays in the legal holiday changes every year, large errors are generated by calculating the periodicity of a flow time sequence according to the natural year, and the calculation and prediction are carried out by combining a crawler system, so that the influence of holiday factors on the flow change can be effectively eliminated, the robustness of the model can be effectively improved, and the prediction accuracy and the application range of the model can be improved.
In another embodiment, step S13 specifically includes:
training a pre-established machine learning model according to the feature set to obtain a flow prediction model, which specifically comprises the following steps:
taking the user data of the preset date as a training sample;
inputting the feature set into a pre-established machine learning model, and training the machine learning model according to the training sample;
and performing parameter optimization on the machine learning model obtained by training by adopting the training sample to obtain a flow prediction model.
It can be understood that the feature set obtained in S12 is used as a feature of a training model, the user data on a preset date is used as a training sample, the real flow value on the preset date is used as a label of the training sample, and a preset machine learning model, such as an SVM support vector machine model, is used for training.
The kernel function uses a linear kernel function, values of parameters such as gamma and cost of the SVM model are traversed by the training samples, the prediction result of the flow prediction model and the absolute error rate of the real flow value are calculated, and the optimal parameters are selected. Namely, the prediction effect of the flow prediction model can be evaluated by determining the error rate so as to adjust the model parameters and effectively improve the prediction accuracy of the model.
The absolute error rate calculation formula is as follows:
further, a date to be predicted is input, and the flow value of the current day is predicted by adopting the flow prediction model.
In the embodiment, a hybrid model of a Holt-Winters cubic exponential smoothing algorithm and an SVM support vector machine algorithm is innovatively adopted, so that the model can process characteristics of more dimensions, the prediction period can be effectively prolonged, the flow can be predicted for a long time in advance, and a model optimization method is provided, so that the effect of model optimization can be effectively improved.
Referring to fig. 3, which is a schematic structural diagram of a mobile communication network traffic prediction apparatus according to an embodiment of the present invention, including:
the data acquisition module 1 is used for acquiring the filling data by adopting a preset filling data prediction model according to the acquired user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number;
the feature acquisition module 2 is used for obtaining a feature set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data;
the model construction module 3 is used for training a pre-established machine learning model according to the feature set to obtain a flow prediction model;
and the flow prediction module 4 is used for inputting the date to be predicted into the flow prediction model to obtain a flow prediction value.
Preferably, the mobile communication network traffic prediction apparatus includes a dividing module for the user data of the preset date:
the user data dividing unit is used for dividing the collected user data of the preset date into a holiday sequence and a natural day sequence;
the system comprises a holiday sequence dividing unit, a holiday sequence calculating unit and a holiday user number time sequence calculating unit, wherein the holiday sequence dividing unit is used for dividing the holiday sequence into a holiday flow time sequence and a holiday user number time sequence;
and the natural day sequence dividing unit is used for dividing the natural day sequence into a natural day traffic time sequence and a natural day user number time sequence.
Preferably, the mobile communication network traffic prediction apparatus includes a building module of the padding data prediction model:
a filling data prediction model unit for filling data prediction model including a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model
The holiday flow prediction model construction unit is used for training a pre-established machine learning model according to the holiday flow time sequence to obtain a holiday flow prediction model;
the holiday user number prediction model building unit is used for training the machine learning model according to the holiday user number time sequence to obtain a holiday user number prediction model;
the natural daily traffic prediction model construction unit is used for training the machine learning model according to the natural daily traffic time sequence to obtain a natural daily traffic prediction model;
and the natural daily user number prediction model construction unit is used for training the machine learning model according to the natural daily user number time sequence to obtain a natural daily user number prediction model.
Preferably, the data acquisition module 1 comprises:
the judging unit is used for judging the type of the date to be predicted by adopting a crawler system;
the date threshold setting unit to be filled is used for determining the date threshold to be filled according to the type of the date to be predicted;
and the filling data prediction unit is used for inputting the threshold value of the date to be filled into a filling data prediction model corresponding to the type of the date to be predicted for prediction to obtain filling data.
In an alternative embodiment, the padding data prediction unit comprises:
a holiday to-be-filled date threshold setting unit, configured to, when it is determined that the date to be predicted is a holiday, take a date set of all holidays in a time threshold between the preset date and the date to be predicted as the date to be filled;
and the holiday filling data prediction unit is used for inputting the date threshold to be filled into the holiday flow prediction model and the holiday user number prediction model to obtain filling data.
Preferably, the feature acquisition module 2 includes:
the calculating unit is used for calculating the related characteristic values of the flow value and the user quantity according to the filling data;
the external data acquisition unit is used for crawling the type of the date to be predicted and the weather data as external data according to a crawler system;
and the characteristic set generating unit is used for generating a characteristic set according to the relevant characteristic value and the external data.
Preferably, the model building module 3 comprises:
the training sample setting unit is used for taking the user data of the preset date as a training sample;
the training unit is used for inputting the feature set into a pre-established machine learning model and training the machine learning model according to the training sample;
and the parameter tuning unit is used for performing parameter tuning on the machine learning model obtained by training by adopting the training sample to obtain a flow prediction model.
Fig. 3 is a schematic structural diagram of a mobile communication network traffic prediction apparatus according to an embodiment of the present invention. The mobile communication network traffic prediction apparatus of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the embodiments of the traffic prediction method for the mobile communication network are implemented, for example, step S13 in fig. 1 trains a pre-established machine learning model according to the feature set to obtain a traffic prediction model. Or, the processor implements the functions of each module/unit in the above device embodiments when executing the computer program, for example, the model construction module 3 is configured to train a pre-established machine learning model according to the feature set to obtain a flow prediction model.
Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory and executed by the processor to implement the invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the mobile communication network traffic prediction device.
The mobile communication network flow prediction device can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The mobile communication network traffic prediction device may include, but is not limited to, a processor, a memory. It will be understood by those skilled in the art that the schematic diagram is merely an example of the mobile communication network traffic prediction apparatus, and does not constitute a limitation of the mobile communication network traffic prediction apparatus, and may include more or less components than those shown, or combine some components, or different components, for example, the mobile communication network traffic prediction apparatus may further include an input and output device, a network access device, a bus, etc.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor is a control center of the mobile communication network traffic prediction device, and various interfaces and lines are used to connect various parts of the whole mobile communication network traffic prediction device.
The memory may be used for storing the computer programs and/or modules, and the processor may implement various functions of the mobile communication network traffic prediction apparatus by executing or executing the computer programs and/or modules stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Wherein, the module/unit integrated with the mobile communication network traffic prediction device can be stored in a computer readable storage medium if it is implemented in the form of software functional unit and sold or used as a stand-alone product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple 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. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (6)
1. A method for predicting traffic of a mobile communication network is characterized by comprising the following steps:
acquiring filling data by adopting a preset filling data prediction model according to the collected user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number;
obtaining a feature set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data;
training a pre-established machine learning model according to the feature set to obtain a flow prediction model;
inputting the date to be predicted into the flow prediction model to obtain a flow prediction value;
the method further comprises the step of dividing the user data of the preset date according to the type of the date, wherein the type of the date comprises holidays and natural days, and the holidays are legal holidays and comprise New year, spring festival, Qingming festival, labor festival, early afternoon festival, mid-autumn festival and national celebration festival; the natural day is a normal working day and a double holiday after the holiday date is removed, and the step of dividing the user data of the preset date according to the type of the date specifically comprises the following steps:
dividing the collected user data of the preset date into a holiday sequence and a natural day sequence;
dividing the holiday sequence into a holiday flow time sequence and a holiday user number time sequence;
dividing the natural day sequence into a natural day traffic time sequence and a natural day user number time sequence;
wherein the method further comprises the step of constructing the filling data prediction model:
the filling data prediction model comprises a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model;
training a pre-established machine learning algorithm according to the holiday flow time sequence to obtain a holiday flow prediction model;
training the machine learning algorithm according to the holiday user number time sequence to obtain a holiday user number prediction model;
training the machine learning algorithm according to the natural daily traffic time sequence to obtain a natural daily traffic prediction model;
training the machine learning algorithm according to the natural daily user number time sequence to obtain a natural daily user number prediction model;
the method for acquiring the filling data by adopting the preset filling data prediction model according to the collected user data on the preset date specifically comprises the following steps:
judging the type of the date to be predicted by adopting a crawler system;
determining a date threshold to be filled according to the type of the date to be predicted;
inputting the threshold value of the date to be filled into a filling data prediction model corresponding to the type of the date to be predicted for prediction to obtain filling data;
wherein, according to the filling data and the external data, a feature set is obtained, specifically:
calculating the related characteristic values of the flow value and the user quantity according to the filling data;
crawling the type of the date to be predicted and weather data as external data according to a crawler system;
and generating a feature set according to the relevant feature value and the external data.
2. The method for predicting mobile communication network traffic according to claim 1, wherein the step of inputting the threshold value of the date to be padded into the filling data prediction model corresponding to the type of the date to be predicted to perform prediction to obtain filling data specifically comprises:
when the date to be predicted is judged to be the holiday date, taking the date set of all holidays in the time threshold value between the preset date and the date to be predicted as the date threshold value to be filled;
and inputting the threshold value of the date to be filled into the holiday flow prediction model and the holiday user number prediction model to obtain filling data.
3. The method for predicting traffic in a mobile communication network according to claim 1, wherein the training of the pre-established machine learning model according to the feature set is performed to obtain a traffic prediction model, and specifically comprises:
taking the user data of the preset date as a training sample;
inputting the feature set into a pre-established machine learning model, and training the machine learning model according to the training sample;
and performing parameter optimization on the machine learning model obtained by training by adopting the training sample to obtain a flow prediction model.
4. A mobile communication network traffic prediction apparatus, comprising:
the data acquisition module is used for acquiring the filling data by adopting a preset filling data prediction model according to the acquired user data on the preset date; the filling data is user data of a time threshold value between a preset date and a date to be predicted, and the user data is time sequence data of a flow value and a user number;
the characteristic acquisition module is used for obtaining a characteristic set according to the filling data and the external data; wherein the external data comprises the type of the date to be predicted and weather data;
the model construction module is used for training a pre-established machine learning model according to the feature set to obtain a flow prediction model;
the flow prediction module is used for inputting the date to be predicted into the flow prediction model to obtain a flow prediction value;
the data acquisition module is further used for executing the step of dividing the user data of the preset date according to the type of the date, and the type of the date comprises a holiday and a natural day; the holidays are legal holidays and comprise New year, spring festival, Qingming festival, labor festival, early afternoon festival, mid-autumn festival and national celebration festival; the natural day is a normal working day and a double holiday after the holiday date is removed:
the step of dividing the user data of the preset date according to the type of the date specifically includes:
dividing the collected user data of the preset date into a holiday sequence and a natural day sequence;
dividing the holiday sequence into a holiday flow time sequence and a holiday user number time sequence;
dividing the natural day sequence into a natural day traffic time sequence and a natural day user number time sequence;
the data acquisition module is further configured to perform a step of constructing a padding data prediction model, where the step of constructing the padding data prediction model specifically includes:
the filling data prediction model comprises a holiday flow prediction model, a holiday user number prediction model, a natural day flow prediction model and a natural day user number prediction model;
training a pre-established machine learning algorithm according to the holiday flow time sequence to obtain a holiday flow prediction model;
training the machine learning algorithm according to the holiday user number time sequence to obtain a holiday user number prediction model;
training the machine learning algorithm according to the natural daily traffic time sequence to obtain a natural daily traffic prediction model;
training the machine learning algorithm according to the natural daily user number time sequence to obtain a natural daily user number prediction model;
the data obtaining module is specifically configured to:
judging the type of the date to be predicted by adopting a crawler system;
determining a date threshold to be filled according to the type of the date to be predicted;
inputting the threshold value of the date to be filled into a filling data prediction model corresponding to the type of the date to be predicted for prediction to obtain filling data;
the feature acquisition module is specifically configured to:
calculating the related characteristic values of the flow value and the user quantity according to the filling data;
crawling the type of the date to be predicted and weather data as external data according to a crawler system;
and generating a feature set according to the relevant feature value and the external data.
5. A mobile communications network traffic prediction apparatus comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the mobile communications network traffic prediction method of any one of claims 1 to 3 when executing the computer program.
6. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the method for traffic prediction in a mobile communication network according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811543654.3A CN109495318B (en) | 2018-12-17 | 2018-12-17 | Mobile communication network flow prediction method, device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811543654.3A CN109495318B (en) | 2018-12-17 | 2018-12-17 | Mobile communication network flow prediction method, device and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109495318A CN109495318A (en) | 2019-03-19 |
| CN109495318B true CN109495318B (en) | 2021-11-16 |
Family
ID=65710555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811543654.3A Active CN109495318B (en) | 2018-12-17 | 2018-12-17 | Mobile communication network flow prediction method, device and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109495318B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832780A (en) * | 2019-04-15 | 2020-10-27 | 成都鼎桥通信技术有限公司 | Private network data prediction method, private network data prediction device, private network data prediction equipment and storage medium |
| CN110175711A (en) * | 2019-05-17 | 2019-08-27 | 北京市天元网络技术股份有限公司 | One kind being based on joint LSTM base station cell method for predicting and device |
| CN112202585B (en) * | 2019-07-08 | 2022-07-01 | 中国移动通信集团浙江有限公司 | Wireless network flow prediction method and device and electronic equipment |
| CN110474808B (en) * | 2019-08-20 | 2022-02-18 | 中国联合网络通信集团有限公司 | Flow prediction method and device |
| CN110750728A (en) * | 2019-10-28 | 2020-02-04 | 精硕科技（北京）股份有限公司 | Prediction method and prediction device for browsing resource heat and readable storage medium |
| CN110880987A (en) * | 2019-11-07 | 2020-03-13 | 中通服咨询设计研究院有限公司 | Packet domain network capacity prediction method and system based on time characteristics |
| CN113015116B (en) * | 2019-12-20 | 2022-06-03 | 中国移动通信集团陕西有限公司 | Dynamic quota method and device based on flow prediction |
| CN111429180A (en) * | 2020-03-23 | 2020-07-17 | 上海中旖能源科技有限公司 | Liquefied natural gas demand prediction method and device |
| CN113542046B (en) * | 2020-04-21 | 2023-01-10 | 百度在线网络技术（北京）有限公司 | Flow estimation method, device, equipment and storage medium |
| CN111769985A (en) * | 2020-06-29 | 2020-10-13 | 中国联合网络通信集团有限公司 | Data traffic prediction method and device |
| CN113973339B (en) * | 2020-07-10 | 2023-04-18 | 中国移动通信集团重庆有限公司 | Load coordination method and device for 4G cell and 5G cell and computer storage medium |
| US11669751B2 (en) * | 2020-11-27 | 2023-06-06 | At&T Intellectual Property I, L.P. | Prediction of network events via rule set representations of machine learning models |
| CN113259158B (en) * | 2021-05-11 | 2022-12-27 | 北京沃东天骏信息技术有限公司 | Network flow prediction method and equipment, model construction and training method and device |
| CN114500231B (en) * | 2022-01-20 | 2023-07-25 | 中通服创立信息科技有限责任公司 | Early warning method and system for flow fluctuation, electronic equipment and storage medium |
| CN115802366B (en) * | 2023-02-13 | 2023-04-28 | 网络通信与安全紫金山实验室 | Network traffic prediction method, device, computer equipment and medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932264A (en) * | 2012-11-02 | 2013-02-13 | 曙光云计算技术有限公司 | Method and device for judging flow overflowing |
| CN104394538A (en) * | 2014-11-28 | 2015-03-04 | 重庆大学 | Mobile network data flow analysis and prediction method |
| CN105515857A (en) * | 2015-12-08 | 2016-04-20 | 江苏省邮电规划设计院有限责任公司 | Metropolitan area data network traffic calculation method based on RBF |
| CN106301984A (en) * | 2015-06-01 | 2017-01-04 | 中国移动通信集团公司 | A kind of mobile communications network capacity prediction methods and device |
| JP2017208757A (en) * | 2016-05-20 | 2017-11-24 | 日本電信電話株式会社 | Device and method for traffic prediction |
| CN108234198A (en) * | 2017-12-19 | 2018-06-29 | 清华大学 | A kind of base station method for predicting and equipment |
| CN108985521A (en) * | 2018-08-02 | 2018-12-11 | 广东工业大学 | Output power predicting method, device, equipment and the medium of photovoltaic generating system |
-
2018
- 2018-12-17 CN CN201811543654.3A patent/CN109495318B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932264A (en) * | 2012-11-02 | 2013-02-13 | 曙光云计算技术有限公司 | Method and device for judging flow overflowing |
| CN104394538A (en) * | 2014-11-28 | 2015-03-04 | 重庆大学 | Mobile network data flow analysis and prediction method |
| CN106301984A (en) * | 2015-06-01 | 2017-01-04 | 中国移动通信集团公司 | A kind of mobile communications network capacity prediction methods and device |
| CN105515857A (en) * | 2015-12-08 | 2016-04-20 | 江苏省邮电规划设计院有限责任公司 | Metropolitan area data network traffic calculation method based on RBF |
| JP2017208757A (en) * | 2016-05-20 | 2017-11-24 | 日本電信電話株式会社 | Device and method for traffic prediction |
| CN108234198A (en) * | 2017-12-19 | 2018-06-29 | 清华大学 | A kind of base station method for predicting and equipment |
| CN108985521A (en) * | 2018-08-02 | 2018-12-11 | 广东工业大学 | Output power predicting method, device, equipment and the medium of photovoltaic generating system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109495318A (en) | 2019-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109495318B (en) | Mobile communication network flow prediction method, device and readable storage medium | |
| CN109872535B (en) | Intelligent traffic passage prediction method, device and server | |
| CN103778474A (en) | Resource load capacity prediction method, analysis prediction system and service operation monitoring system | |
| EP3961384A1 (en) | Automatic derivation of software engineering artifact attributes from product or service development concepts | |
| CN112508118A (en) | Target object behavior prediction method aiming at data migration and related equipment thereof | |
| CN113486584B (en) | Method and device for predicting equipment failure, computer equipment and computer readable storage medium | |
| CN108875048B (en) | Report generation method and device, electronic equipment and readable storage medium | |
| CN115545331A (en) | Control strategy prediction method and device, equipment and storage medium | |
| CN111800807A (en) | Method and device for alarming number of base station users | |
| CN113837383A (en) | Model training method and device, electronic equipment and storage medium | |
| CN112785089A (en) | Agent service configuration method and device, electronic equipment and storage medium | |
| CN109978172B (en) | Resource pool utilization rate prediction method and device based on extreme learning machine | |
| CN113343577B (en) | Parameter optimization method, device, equipment and medium based on machine learning | |
| CN114520773A (en) | Service request response method, device, server and storage medium | |
| CN113763053A (en) | Block chain-based method for evaluating user activity level of big data e-commerce platform | |
| CN111339156B (en) | Method, apparatus and computer readable storage medium for long-term determination of business data | |
| CN112085497A (en) | User account data processing method and device | |
| CN108629610B (en) | Method and device for determining popularization information exposure | |
| CN111783487A (en) | Fault early warning method and device for card reader equipment | |
| CN113128734A (en) | Method and device for predicting oil field yield | |
| CN111104569A (en) | Region segmentation method and device for database table and storage medium | |
| CN114338429B (en) | Network bandwidth determining method and device and electronic equipment | |
| CN108804640B (en) | Data grouping method, device, storage medium and equipment based on maximized IV | |
| CN116822997B (en) | Real-time carbon emission accounting method and device considering green power | |
| Shah | An Exogenous Factor Aware Resource Prediction Model for Auto-Scaling in Cloud |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB02 | Change of applicant information |
Address after: 510665 East Floor 3, No. 14 and No. 16 Jianzhong Road, Tianhe District, Guangzhou City, Guangdong Province Applicant after: Yitong Century Science and Technology Co., Ltd. Address before: 510665 12/F, Building A, Guangzhou Information Port, 16 Keyun Road, Tianhe District, Guangzhou City, Guangdong Province Applicant before: Guangdong Eastone Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |