TWI522959B - Mobile network automation traffic growth prediction system - Google Patents

Description

Mobile network automation traffic growth prediction system

The present invention relates to a traffic growth prediction system; in particular, a mobile network automation service growth prediction system, which collects and analyzes various spaces on a mobile network periodically through an automated management mechanism of a smart hub. The amount of traffic and data users of each terminal device under the segment, through the logistic equation, predicts the growth trend of future traffic and data users, and presents the results through the human-machine interface. The intelligent hub also has a smart predictive saturation parameter tuning mechanism, which can automatically adjust the parameters based on the gap between the actual traffic data and the predicted results periodically or irregularly, continuously reduce the prediction error, and improve the overall network traffic. The accuracy of the forecast.

With the development of mobile communication technologies, the increasing complexity of mobile network architectures and functions, and the explosive growth of mobile network traffic, today's mobile network capacity has reached bottlenecks more quickly, and in the past it was only for specific mobile network devices. Traffic or traffic monitoring, if the monitoring threshold is reached, then expand the specific mobile network device. However, due to the limited wireless transmission resources of mobile network devices, it is not the same as the wired network, as long as the network equipment that reaches the bottleneck threshold is expanded. For example, in the case of mobile network base stations, increasing the space density of a large number of stations is not only costly, but also causes mutual interference between stations. Blindly amplifying mobile network equipment cannot effectively solve network capacity bottlenecks. problem.

In addition to the current forecasting technology, the most similar patent is the Chinese patent number: 201010300133, which is mainly for the mobile network traffic with the SAS commercial software system for the growth trend mathematical model fitting to predict, and its SAS system is a commercial software system, its construction The cost is high, and it is not easy to use free software construction and implementation. In addition, its mechanism does not have the mechanism to automatically update the forecast results regularly, and it cannot reflect the current status of the mobile network in a timely and realistic manner. The mechanism does not have a mechanism to compare the predicted value with the actual value of the current state of the network. It is easy to cause the prediction result to reflect the current situation of the network when the time and network variation are large. And its prediction is only for the traffic prediction in the space of the mobile network, and it is impossible to use the mobile device and its related data users and single-person traffic usage for each terminal device (such as smart phone, tablet). Forecast.

It can be seen that there are still many shortcomings in the above-mentioned methods of use, which is not a good design, but needs to be improved.

In view of the shortcomings derived from the above-mentioned conventional methods, the inventor of the present invention has improved and innovated, and after years of painstaking research, he finally succeeded in researching and developing this invention.

The object of the present invention is to provide a mobile network automatic service growth prediction system, which can automatically and periodically collect mobile network traffic and configuration data and perform translation analysis to perform terminal device services in each space. Forecast of future growth trends of data users and single-person usage, presented by human-machine interface To grow the trend results, and with intelligent prediction parameter saturation value adjustment mechanism, the prediction error can be dynamically adjusted to make the prediction result approximate the current state of the mobile network. The results of the mobile network traffic growth trend forecast will provide an important reference for the construction of mobile networks. The network capacity of the mobile network required for each year will be planned from the perspective of the overall network, and the deployment of mobile network expansion will be carried out in advance. Optimize the performance of your network by understanding network load.

The mobile network automation service growth prediction system for achieving the above object aims to collect the traffic data of the mobile network interface and the configuration data of each network component by using the front-end network component and the interface extractor through the translator. Statistics and analysis of the amount of traffic and data users of each terminal device in each space segment of the mobile network, store it in a database, and then predict the space segments by logistic equations through the back-end predictive computing unit. The growth trend of the number of messages and users of each terminal device. Multiplying the number of data users by the average traffic usage per person is the traffic growth trend of each terminal device in each space segment. The sum of the terminal devices under each space segment is the space of the mobile network. The overall growth trend of the segment under the segment, and then store the predicted results in the database.

The human-machine interface obtains the predicted results through the database, and displays the results in a graphical interface to predict the growth trend of mobile network traffic, data users and average traffic usage per terminal device in each space segment in the next few years. And the total traffic ratio map of each terminal device under each space segment. The intelligent hub has a smart saturation parameter tuning mechanism, which can automatically adjust the parameters based on the gap between the actual traffic data and the prediction result periodically or irregularly, continuously reduce the prediction error, and improve the accuracy of the overall network traffic prediction. Degree, making its predictions closer to the current state of the mobile network. The predictive analyzer then counts the predicted results and analyzes and statistically notes and alerts for the set traffic threshold. The smart hub is responsible for managing network components and interface extractors, translators, databases, predictive operators, Predictive analyzer's automated workflow, condition monitoring and alerting mechanisms.

When the patented technical features provided by the present invention are compared with other conventional technologies, the following advantages are obtained:

1. In addition to monitoring the traffic information of each space segment of the mobile network, the present invention can also predict the annual growth trend of the traffic in the future.

2. In addition to monitoring the number of data users in each space segment of the mobile network, the present invention can also predict the growth trend of the number of data users in the future.

3. In addition to monitoring the average traffic usage per person in each space segment of the mobile network, the present invention can also predict the growth trend of average per-person traffic usage per person in the future.

4. In addition to monitoring the traffic information of each terminal device (such as smart phone, tablet, network card, function mobile phone, etc.) in each space segment of the mobile network, the present invention can also predict the annual device of each terminal device in each space segment in the future. The growth trend of the news.

5. The present invention can monitor the number of data users of each terminal device (such as smart phone, tablet, network card, function mobile phone, etc.) in each space segment of the mobile network, and can also predict each terminal device in each space segment in the future. The number of data users in the year is growing.

6. In addition to monitoring the average traffic usage information of each terminal device (such as smart phone, tablet, network card, function mobile phone, etc.) in each space segment of the mobile network, the present invention can also predict future space segments. The average daily usage of each terminal device grows.

7. The present invention can update the prediction results periodically (e.g., weekly) through system automation to update the prediction results in real time in response to changes in the current state of the mobile network.

8. The present invention can present actual traffic change trends and predictive messages through a human-machine interface. Graphical comparison chart of the trend of change, for system users to compare the accuracy of the estimation operation in time.

9. The invention can automatically adjust the parameters according to the gap between the actual traffic data and the prediction result through a smart hub periodically or irregularly, continuously reduce the prediction error, improve the accuracy of the overall traffic prediction, and make predictions. The result is closer to the current state of the mobile network.

11‧‧‧Mobile Network

12‧‧‧Data Network

21‧‧‧Network components and interface extractors

21a‧‧‧ Data Capture Module

22b‧‧‧Data Acquisition Management Module

22‧‧‧Translator

22a‧‧‧Data Analysis Module

22b‧‧‧Data Analysis Management Module

23‧‧‧ Database

23a‧‧‧ Data Storage Module

23b‧‧‧Database Management Module

24‧‧‧Predictor

24a‧‧‧Predictive Computing Module

24b‧‧‧Predictive Computing Management Module

25‧‧‧Predictive Analyzer

25a‧‧‧Predictive Analysis Module

25b‧‧‧Predictive Analysis Management Module

26‧‧‧Human Machine Interface

26a‧‧‧Human Machine Interface Presentation Module

26b‧‧‧Human Machine Interface Management Module

27‧‧‧Smart Hub

27a‧‧‧System Management Human Machine Interface

27b‧‧‧System Management Module

31‧‧‧System users

32‧‧‧System Manager

241‧‧‧Data analysis unit

242‧‧‧Linear unit

243‧‧‧predictive arithmetic unit

243a‧‧ ‧ Logistic unit

243b‧‧‧Backward Logistic Unit

Please refer to the detailed description of the present invention and its accompanying drawings, and the technical contents of the present invention and the functions thereof can be further understood. The related drawings are: FIG. 1 is the structure of the mobile network automatic service growth prediction system of the present invention. Figure.

FIG. 2 is a system module diagram of the mobile network automatic service growth prediction system of the present invention.

FIG. 3 is a block diagram of a prediction operation module of a predictive computing unit of the mobile network automation service growth prediction system of the present invention.

FIG. 4 is a flow chart diagram of the smart predictive tuning hub module of the intelligent hub of the mobile network automatic traffic growth prediction system of the present invention.

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Hereinafter, the present invention will be further described with reference to the accompanying drawings: the present invention is a mobile network automated service growth prediction system.

The invention is provided with automated collection of mobile network traffic data, traffic data analysis and statistics, traffic and data user number prediction, human-machine interface presentation of graphical traffic growth trend results, and human-machine interface dynamic adjustment prediction parameters. Mobile network automation traffic growth prediction system.

Please refer to FIG. 1. FIG. 1 is a structural diagram of a mobile network automation service growth prediction system according to the present invention. As shown in FIG. 1 , the network interface in the mobile network 11 is the data source object of the mobile network automatic traffic growth prediction system of the present invention, and the network component and interface extractor 21 and the translator 22 The database 23, the predictive arithmetic unit 24, the predictive analyzer 25, the human machine interface 26, and the smart hub 27 are constituent elements of the present invention. The components of the present invention all have data network networking functions and are all connected to the data network 12 to achieve the automation goal of mobile network traffic growth prediction. The human-machine interface 26 can be a website interface, a smart phone interface or a Windows Form interface, and the data network or mobile network connection method can be used to obtain the mobile network traffic growth prediction result in the database 23, and the system user 31 can The human machine interface 26 queries the prediction results.

One or more network elements and interface extractor 21 are connected to the mobile network 11 through the data network, and the network packet data of the network interface of the mobile network 11 is periodically collected in an automated manner, and the packet is parsed and parsed. In operation, the data in the packet is retrieved and transmitted to the translator 22 via the data network 12, and the network component and interface extractor 21 are also obtained from the network components of the mobile network 11. The configuration data is transferred to the translator 22. The translator 22 parses and translates the traffic and the configuration data, and generates statistical values of the terminal device traffic and the number of data users in each space segment, and stores the data in the database 23. The predictive computing unit 24 obtains the traffic volume, the number of data users, and the logistic prediction operation parameters of each terminal device in each space segment through the data network 12, and then performs logistic prediction operation. The process further generates future growth trend data of the traffic volume, the number of data users, and the average traffic usage of each terminal device in each space segment, and then stores the result in the database 23 through the data network 12. The prediction analyzer 25 obtains the growth trend data of the future traffic, the number of data users, and the data usage of each terminal device in each space segment through the data network 12 to the database 23, and performs data statistics and statistical result alarm operations. The statistics and alarm results are then stored in the database 23 via the data network 12. The human machine interface 26 obtains the growth trend data of the future traffic, the number of data users, and the average traffic usage of each terminal device in each space segment through the data network 12 to the database 23, and the statistics and alarm results of the data. The data is presented in a human-machine interface through graphical technology for user inquiry. The human-machine interface 26 also provides logistic prediction operation parameter query and setting interface, and provides system user 31 to query and edit logistic (logistic). Predict the parameters required for the operation.

The results of the Smart Center 27 Sustainable View Database 23 generate periodic, (eg daily, weekly, monthly, quarterly and yearly) analysis reports of the Mobile Network Automation Traffic Growth Prediction System to the System Manager 32, and When a system operation problem is discovered, an alert is transmitted to the system administrator 32.

The components of the present invention are controlled by the smart hub 27, managing system components (eg, adding or removing components), setting system parameters (eg, setting network components and interface extractor 21 to periodically collect mobile networks) Service and group State data), setting network component and interface extractor 21, translator 22, database 23, predictive computing unit 24 and predictive analyzer 25 automated processes, periodically detecting system component anomalies, such as viewing results in database 23, periodically (for example, daily, weekly, monthly, quarterly, and yearly) generates a status analysis report of the mobile network automation service growth prediction system to the system administrator 32, and transmits an alarm to the system administrator when the system operation problem is discovered. 32. The system manager 32 can perform the management mechanism of each unit through the smart hub 27.

The intelligent hub 27 also has a smart predictive saturation parameter tuning mechanism, which can periodically compare the actual and predicted values of the current state of the network, and dynamically adjust the saturation value parameters to adjust the prediction results so that the prediction results are closer to the current state of the mobile network.

The network element and interface extractor 21, the translator 22, the database 23, the predictive computing unit 24 and the predictive analyzer 25, the smart hub 27, and the human machine interface 26 can all be configured as one or more hardware platforms. The system capacity plan is large or small, and whether system component backup is needed. It is also possible to implement some (or even all) components on the same hardware platform.

Please refer to FIG. 2, which is a system module diagram of the mobile network automatic service growth prediction system of the present invention. As shown in FIG. 2, the network component and interface extractor 21 includes two modules: a data capture module 21a and a data capture management module 21b. The data capture module 21a and the data capture management module Group 21b communicates through an API (Application Programming Interface). The data capture module 21a connects the network components of the mobile network 11 through the data network 12 through technologies such as FTP, SFTP, Telnet, or database connection, and collects configuration data of each network component. The network interface of the mobile network 11 is connected through the data network 12, and the network packet passing through the network interface is obtained, and the packet data analysis operation is performed, and the data online user is captured. Information.

After the data is collected, it is transmitted to the translator 22 via the data network 12 via FTP, SFTP, Telnet, and the like. The data capture management module 21b is responsible for monitoring the status of the network component and the interface extractor 21, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the data capture management module 21b also periodically reports the state parameters of the network component and the interface extractor 21 to the system management module 27b of the smart hub 27 for state management monitoring. The data capture management module 21b is also responsible for managing the automated workflow of the network component and interface extractor 21, and communicating with the system management module 27b of the smart hub 27, according to the automated workflow set by the system management module 27b. Parameters, regular automated workflow.

The interpreter 22 includes two modules: a data analysis module 22a and a data analysis management module 22b. The data analysis module 22a and the data analysis management module 22b communicate with each other through an API (Application Programming Interface). . The data analysis module 22a obtains the configuration and traffic data collected by the network component and the interface extractor 21 through the data network by using FTP, SFTP, Telnet, etc., and the traffic volume of each terminal device in each space segment. The statistics are counted with the number of data users, and the statistical results are stored by the data storage module 23a of the database 23 using the database connection technology. The data analysis management module 22b is responsible for monitoring the status of the translator 22, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the data analysis management module 22b also periodically reports the state parameters of the translator 22 to the system management module 27b of the smart hub 27 for state management monitoring. The data analysis management module 22b is also responsible for managing the automated workflow of the translator 22, communicating with the system management module 27b of the smart hub 27, and periodically performing automation according to the automated workflow parameters set by the system management module 27b. Industry process.

The database 23 includes two modules: a data storage module 23a and a database management module 23b. The data storage module 23a includes functions of data storage and data editing, etc., a translator 22, a prediction operator 24, a prediction analyzer 25, and The human-machine interface 26 needs to perform data storage and editing operations through the data storage module 23a by using the database connection technology. The database management module 23b is responsible for monitoring the status of the database 23, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the database management module 23b also periodically reports the status parameters of the database 23 to the system management module 27b of the smart hub 27 for status management monitoring. The database management module 23b is also responsible for managing the automated workflow of the database 23, communicating with the system management module 27b of the smart hub 27, and periodically performing automated workflows based on the automated workflow parameters set by the system management module 27b.

The predictive computing unit 24 includes two modules, a predictive computing module 24a and a predictive computing management module 24b, and the predictive computing module 24a communicates with the predictive computing management module 24b via an API (Application Programming Interface). The predictive computing module 24a uses the data connection module 23a to retrieve the traffic volume and data user data of each terminal device in each space segment through the data storage module 23a of the database, and through the logistic prediction operation process. Estimate the annual growth trend of each terminal equipment traffic, number of data users and average traffic volume per person in each space segment, and then use the database connection technology to access the data storage module 23a of the database 23 by using the database connection technology. Store data. The predictive operation management module 24b is responsible for monitoring the state of the predictive computing unit 24, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the predictive operation management module 24b also periodically returns the status parameter of the predictive computing unit 24 to the system management module 27b of the smart hub 27. Perform status management monitoring. The predictive computation management module 24b is also responsible for managing the automated workflow of the predictive computing unit 24, communicating with the system management module 27b of the smart hub 27, and periodically performing automated operations based on the automated workflow parameters set by the system management module 27b. Process.

The predictive analyzer 25 includes two modules, a predictive analysis module 25a and a predictive analysis management module 25b. The predictive analysis module 25a and the predictive analysis management module 25b communicate with each other through an API (Application Programming Interface). The predictive analysis module 25a uses the data connection module 23a to retrieve the annual growth trend change data of each terminal device information, the number of data users, and the average amount of traffic per person in each space segment through the data storage module 23a of the database 23. Calculate the statistical data for this forecast data, such as the time in which each terminal device in each space segment reaches the upper threshold of the traffic, the number of data users, and the average traffic volume per person, and then use this statistical value as data. The library connection technology stores data through the data storage module 23a of the database 23. The predictive analysis management module 25b is responsible for monitoring the status of the predictive analyzer 25, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the predictive analysis management module 25b also periodically reports the status parameters of the predictive analyzer 25 to the system management module 27b of the smart hub 27 for status management monitoring. The predictive analysis management module 25b is also responsible for managing the automated workflow of the predictive analyzer 25, communicating with the system management module 27b of the smart hub 27, and performing automated operations periodically according to the automated workflow parameters set by the system management module 27b. Process.

The human-machine interface 26 includes two modules: a human-machine interface presentation module 26a and a human-machine interface management module 26b. The human-machine interface presentation module 26a and the human-machine interface management module 26b are interfaced through the programming interface (API, Application Programming). Interface) to communicate. The human interface presentation module 26a The database connection technology obtains the number of traffic and data users of each terminal device in each space segment, the number of terminal device services in each space segment, the number of data users, and the average per person through the data storage module 23a of the database 23. The annual growth trend of the traffic volume, and the statistics of the time limit of the traffic, the number of data users and the average threshold of the average traffic of each terminal in each space segment, through the graphical The interface method, such as the use of line graphs, pie charts and other graphics, and report format to present data, also provides a query interface for system users 31 to change the query conditions due to their needs, so that system users 31 can respond to the current state of the mobile network Mutation, regularly review the information on the changes in the annual growth trend of each terminal equipment, data users and average traffic per person in each space segment. The human-machine interface management module 26b is responsible for monitoring the state of the human-machine interface 26, and reporting the alarm information to the system management module 27b of the smart hub 27 for alarm processing when an obstacle occurs. At the same time, the human interface management module 26b also periodically reports the status parameters of the human interface 26 to the system management module 27b of the smart hub 27 for status management monitoring.

Each component includes a management module, which is a data acquisition management module 21b, a data analysis management module 22b, a database management module 23b, a predictive computing management module 24b, a predictive analysis management module 25b, and a person. Machine interface management module 26b.

The main functions of the management modules (21b, 22b, 23b, 24b, 25b, 26b) of each component are: (1) Before the components are started, the management modules (21b, 22b, 23b, 24b, which constitute the components, 25b, 26b) upload component parameters and automation operation information to the system management module 27b; (2) before the components start to operate, the component management modules (21b, 22b, 23b, 24b, 25b, 26b) accept the setting commands issued by the system management module 27b, perform parameter setting and automatic operation setting on each component; (3) management modules (21b, 22b, 23b, 24b, of each component component, 25b, 26b) Execute the automated operation procedures of each component according to the completed automation operation setting; (4) After each component starts to operate, the management modules of each component (21b, 22b, 23b, 24b, 25b, 26b) The component status information is transmitted to the system management module 27b at timing and when an abnormal condition is found.

The smart hub 27 includes a system management module 27b and a system management human interface 27a. The system management human interface 27a and the system management module 27b communicate with each other through an API (Application Programming Interface). The system management human interface 27a is mainly provided to the system administrator 32 for system management interface. The main functions of the system management module 27b are:

(1) Add components (21, 22, 23, 24, 25, 26) to the system, or remove them.

(2) Before the components are started, the system management module 27b collects the parameters uploaded by the management modules (21b, 22b, 23b, 24b, 25b, 26b) of each component and the automation operation information, and manages the system. The human machine interface 27a presents information to the system administrator 32.

(3) Before the components are started, the system management module 27b issues the management module (21b, which is set by the system administrator 32 to the system management man-machine interface 27a for various component parameters and automation operations. 22b, 23b, 24b, 25b, 26b).

(4) After each component starts to operate, the system management module 27b collects the constituent elements. The management modules (21b, 22b, 23b, 24b, 25b, 26b) are timed and the component status information uploaded when an abnormal condition is found and processed and presented subsequently. If the system management module 27b finds that the management modules (21b, 22b, 23b, 24b, 25b, 26b) of the respective component elements may not be in an alive state (that is, the component state information is not regularly uploaded) or the components In the event of an abnormal condition, the system management module 27b can issue an alert to the system administrator 32 (eg, a alerts, phone, email, voice, or light).

Please refer to FIG. 3, which is a block diagram of a prediction operation module of a predictive computing unit of the mobile network automatic service growth prediction system of the present invention. As shown in FIG. 3, the prediction operation module includes at least three units of a data analysis unit 241, a linear operation unit 242, and a prediction operation unit 243, and the prediction operation unit 243 includes a logistic operation unit 243a and a counter. Two units are returned to the logistic operation unit 243b.

The data analysis unit 241 performs a data ratio assignment operation in which the terminal type cannot be identified, and first acquires the number of traffic and data users of each terminal device in each space segment by the data storage module 23a of the database 23 by the database connection technology. The terminal devices in each space segment can't identify the data and data user data of the data category of which terminal device. According to the proportion of other terminal traffic and data users, the total number of traffic and total data users cannot be allocated. Identify the traffic and data user data of the terminal device category to the data and data user data of other terminal devices, and then divide each terminal traffic data by the number of data users to obtain the average traffic data per person, and the data The number of users and the average traffic data per person are transmitted to the linear operation unit 242 through an API (Application Programming Interface).

The linear operation unit 242 performs linear algebra operation and linear operation The unit 242 obtains the number of data users and the average traffic data of each person and the corresponding time data from the data analysis unit 241, and performs the following two operation processes:

(A) Substituting the number of data users and their corresponding time data into a linear simultaneous equation to obtain the slope, and the linear equation is as follows: aX+b=Y (1) Bring the data of the user number into the mathematical formula ( 1) Y, the weekly time data is brought into the mathematical formula (1) X, and when the data user data has N pens, the solution of the linear simultaneous equations of the N mathematical formulas (1) is the slope a and the intercept b. . Then, the number of data users, the corresponding time data, and the slope a are transmitted to the prediction operation unit 243 through the program interface.

(B) Substituting the average traffic usage per person and its corresponding time data into a linear simultaneous equation to solve the slope. Bring the average traffic usage data of each person into the Y of the mathematical formula (1), the time data is brought into the X of the mathematical formula (1), and when the average traffic data per person has N strokes, the N mathematical expressions are obtained. The solution of the linear simultaneous equation is the slope a and the intercept b. Then, the average traffic usage per person, the corresponding time data, and the slope a are transmitted to the prediction operation unit 243 through the program interface.

The prediction operation unit 243 first determines the magnitude of the obtained data slope, and if the slope is greater than zero, starts a logistic operation unit 243a, and if the slope is less than or equal to zero, starts the inverse logistic operation unit 243b. The logistic operation unit 243a brings the data obtained from the linear operation unit 242 into a logistic equation solution, and the logistic equation is as follows: Bring the time data of the obtained data into X. If the obtained data is the number of data users, the number of data users will be brought into Y. If the data obtained is the average traffic per person, the average traffic per person will be brought into Y. The N value is obtained by the database connection technology through the data storage module 23a of the database 23 to obtain the saturation value of each terminal stored in the database 23. When there is M data, the M log logical equation is solved in a simultaneous manner. Find a and b. Then, the data storage module 23a of the database 23 obtains the future estimated time data of each terminal stored in the database 23 by the database connection technology, and brings the data into the logistic equation X and saturates each terminal. The value is brought into N, and the obtained a and b are brought into the logistic equation, and Y can be obtained, which is the number of data users and the average traffic data of each terminal predicted in the future, and then each space The number of data users of each terminal device in the segment is multiplied by the average traffic of each person, that is, the traffic of each terminal device in each space segment is obtained, and then the terminal devices under each space segment are added, that is, Get the estimated future growth of the service.

The reverse logistic operation unit 243b, whose input data is the number of data users or the average traffic data per person, first obtains the following three values:

(1) Threshold value: Multiply the maximum value of the input data by 2 times.

(2) Data mirror inversion value: (1) Threshold value minus input data.

(3) Saturation value: The threshold value is reduced by 0.9 times (2) The minimum value of the data mirror inversion value. Then, (2) the data mirror inversion value is taken as the input value Y, and the time data corresponding to Y is taken as X, and (3) the saturated value is brought into the logistic operation unit 243a to obtain the estimated value, and then The (1) threshold value is subtracted from the estimated value, which is the final estimation result obtained by the inverse logistic operation unit 243b.

Please refer to FIG. 4, which is a flow chart diagram of the intelligent predictive tuning hub module of the intelligent hub of the mobile network automatic traffic growth prediction system of the present invention. As shown in FIG. 4, the intelligent predictive tuning process 150 can automatically and periodically adjust the parameters based on the gap between the actual traffic data and the predicted result on a regular or irregular basis, continuously narrowing the prediction error, and improving the overall network traffic prediction. Accuracy. First, the smart predictive tuning start 151 program starts the smart predictive tuning process, and through the actual value and predicted value slope operation 152 program, the database connection technology acquires a certain time segment through the data storage module 23a of the database 23. (For example, three months), the mobile network traffic, the number of data users, and the average traffic status data of each person, calculate the linear slope of the actual value, and then obtain the data through the data storage module 23a of the database 23 by the database connection technology. The behavior of the network traffic, the number of data users, and the average traffic of each person in a time zone (such as three months), calculate the slope of the predicted value, and calculate the accuracy difference as the actual value slope minus the predicted value. The slope, then the slope difference and the accuracy threshold comparison 153 procedure, first obtain the accuracy threshold through the data storage module 23a of the database 23 by the database connection technology, and adjust when the absolute value of the slope difference is greater than the accuracy threshold. The saturation value 154 program calculates the following two values:

(1) Accuracy difference percentage: The accuracy difference is divided by the actual value slope.

(2) Saturated value difference: The saturation value is multiplied by the accuracy difference percentage. When the accuracy difference is greater than zero, the saturation adjustment value is the saturation value plus the saturation value difference. When the accuracy difference is less than zero, the saturation adjustment value is the saturation value minus the saturation value difference. The saturation value adjustment mechanism also has an error prevention mechanism, and has a reasonable upper limit of saturation value and a reasonable lower limit threshold comparison mechanism. When the saturation value adjustment value exceeds the upper and lower thresholds, the saturation value is the threshold value. The upper and lower thresholds are stored in the database 23, and the threshold can be manually adjusted. Then, the database connection technology is used to update the saturation value to the saturation adjustment value through the data storage module 23a of the database 23, and then the 155 process is started. In sequence, through the data connection, the prediction operation unit 24a is started to perform the prediction operation flow, and then returns to the actual value and the predicted value slope comparison 152, the slope difference and the accuracy threshold value comparison 153 program and the adjustment saturation value 154 program, when the accuracy is When the absolute value of the difference is larger than the absolute value of the difference of the previous time, the difference of the saturation value is the difference of the previous saturation value divided by 2, and the value of the saturation value is the value of the previous saturation minus the saturation value. The process is repeated until the slope difference is compared with the accuracy threshold value 153. When the absolute value of the slope difference is less than or equal to the accuracy threshold, the flow proceeds to the training termination 156 procedure, at which time the smart prediction saturation parameter tuning process has been performed. Completed, the actual value and the predicted value are quite similar at this time. Another intelligent central 27 system management man-machine interface 27a also provides an accuracy threshold editing interface, which can be provided to the system administrator, in response to the accuracy requirements, the editing accuracy threshold, the accuracy threshold editing interface, through the data The library connection technology changes the accuracy threshold in the database through the data storage module 23a of the database 23.

The detailed description of the present invention is intended to be illustrative of a preferred embodiment of the invention, and is not intended to limit the scope of the invention. The patent scope of this case.

To sum up, this case is not only innovative in terms of technical thinking, but also has many of the above-mentioned functions that are not in the traditional methods of the past. It has fully complied with the statutory invention patent requirements of novelty and progressiveness, and applied for it according to law. Approved this invention patent application, in order to invent invention, to the sense of virtue.

11‧‧‧Mobile Network

12‧‧‧Data Network

21‧‧‧Network components and interface extractors

22‧‧‧Translator

23‧‧‧ Database

24‧‧‧Predictor

25‧‧‧Predictive Analyzer

26‧‧‧Human Machine Interface

27‧‧‧Smart Hub

31‧‧‧System users

32‧‧‧System Manager

Claims (22)

A mobile network automated service growth prediction system, which is connected through a data network, and includes at least: a predictive computing unit, a predictive computing process for mobile network traffic; and a predictive analyzer for mobile network traffic estimation The result is analyzed and statistically operated; a human-machine interface is connected with a system user to provide the function of querying the mobile network traffic prediction result; a smart hub, built-in prediction parameter saturation value adjustment mechanism, and managing each component for The prediction result is a smart adjustment mechanism; a prediction operator receives data and transmits the data to a database storage; a plurality of databases receives the data storage and editing; a network component and interface extractor, through the network The road receives the data and transmits it to a translator; a network component and an interface extractor are connected to the network to collect data; and a translator uses the network component and the interface extractor to collect data and translate the data. Storing to the database; wherein the predictive computing unit further comprises a predictive computing module and a predictive computing management module; and wherein Measuring calculation module further comprises a data analyzing unit is connected, the linear prediction operation unit and an arithmetic unit, and the interface for receiving data transmission programmatically. For example, the mobile network automation service growth prediction system described in claim 1 is characterized in that the main management data of the intelligent hub is a periodic automation, status monitoring or alarm operation process of each component. For example, the mobile network automation service growth prediction system described in claim 1 is characterized in that the main computing data of the predictive computing unit is mobile network traffic, data user number or single person average traffic data. For example, the mobile network automation service growth prediction system described in claim 1 includes the system management module and the intelligent predictive adjustment center module, and each module is implemented through a program interface. communication. For example, the mobile network automation service growth prediction system described in claim 4, wherein the intelligent prediction adjustment center module further includes a calibration mechanism for predicting parameter saturation values, which is dynamically or periodically based on actual or irregular conditions. The gap between the data and the prediction result is automatically adjusted to the saturation value parameter, and the actual value and the predicted value gap are continuously narrowed to improve the accuracy of the prediction. For example, the mobile network automation service growth prediction system described in claim 4, wherein the system management module process further includes: (1) built-in new or removed components mechanism; (2) components Before the components start to operate, collect the parameters and automation operation information uploaded by the management modules of each component, and present the information to the system administrator in the system management man-machine interface; (3) release the system administrator before the components start to operate. In the system management man-machine interface, face the various parameters and the settings completed by the automation operation to the management modules of each component; (4) After the components are started, collect the management modules of each component. And the component status information uploaded when an abnormal condition is found and subsequently processed and presented; and (5) when the management module of each component is found to be in a viable state or an abnormal condition occurs in each component, the system administrator issues Alarm. The mobile network automation service growth prediction system described in claim 1, wherein the predictive computing module process further comprises: (1) performing data storage and retrieval mechanism through a database connection technology to a database; And (2) through the logistic prediction process, the mechanism of the annual growth trend of each terminal device traffic, data users and average traffic volume per person in each space segment is obtained. The mobile network automation service growth prediction system according to claim 1, wherein the predictive operation management module process further comprises: (1) executing an automatic operation program of the predictive computing unit; and (2) monitoring the predictive computing unit. Operational status; and (3) alerting the intelligent hub to the operational status when the abnormal condition is discovered periodically. The mobile network automation service growth prediction system according to claim 1, wherein the prediction operation unit further comprises a logistic operation unit and a reverse logistic operation unit. For example, the mobile network automation service growth prediction system described in claim 1 is characterized in that the data analysis unit has a built-in data proportion distribution mechanism that cannot identify the terminal category. For example, the mobile network automation service growth prediction system described in claim 1 is characterized in that the linear operation unit has a built-in mechanism for linearly calculating data to obtain a growth slope thereof. The mobile network automation service growth prediction system according to claim 9, wherein the logistic operation unit internally builds and integrates multiple data through a logistic process. The mechanism by which logistic equations are evaluated. The mobile network automation service growth prediction system according to claim 9, wherein the reverse logistic operation unit has built-in data mirroring inversion, saturation value or threshold. The mechanism of value. For example, the mobile network automation service growth prediction system described in claim 1 includes the data storage module and the database management module, and transmits and receives data through the program interface. The mobile network automation service growth prediction system according to claim 14, wherein the data storage module further comprises a data storage and data editing mechanism, and the prediction computing device, the prediction analyzer, and the The human-machine interface uses data connection technology for data storage and editing. The mobile network automation service growth prediction system according to claim 14, wherein the database management module process further comprises: (1) executing a database automation operation program; (2) monitoring a database operation state; and 3) Regularly report the operational status to the intelligent hub when an abnormal situation is discovered. For example, the mobile network automation service growth plan described in item 1 of the patent application scope The measurement system, wherein the network component and the interface extractor further comprise a data capture module and a data capture management module, and transmit and receive data through the program interface. The mobile network automation service growth prediction system according to claim 17, wherein the data acquisition module process further comprises: (1) collecting configuration data of each network component and data of a network interface. The mechanism for the user's traffic and user data; and (2) the data storage mechanism through the database connection technology to the database. For example, the mobile network automation service growth prediction system described in claim 17 of the patent application, wherein the data acquisition management module process further comprises: (1) executing an automatic operation procedure of the network component and the interface extractor; 2) monitoring the operation status of the network component and the interface extractor; and (3) alerting the intelligent hub to the operational status when the abnormal condition is found periodically. For example, the mobile network automated service growth prediction system described in claim 1 includes the data analysis module and the data analysis management module, and transmits and receives data through the program interface. For example, the mobile network automation service growth prediction system described in claim 20, wherein the data analysis module process further comprises: (1) configuring the mobile network configuration and the traffic data under each space segment. The statistical analysis mechanism of each terminal device; and (2) the data storage mechanism through the database connection technology to the database. As for the mobile network automation service growth plan described in item 20 of the patent application scope The measurement system, wherein the data analysis management module process further comprises: (1) executing an automatic operation program of the translator; (2) monitoring the operational status of the translator; and (3) reporting the operational status to the intelligent type upon regular occurrence and abnormality detection. The hub performs an alarm.