TWI572179B - Mobile network base station communication prediction and resource scheduling automation analysis system and method - Google Patents

Description

Mobile network base station traffic prediction and resource scheduling automatic analysis system and method

The invention relates to the design of an automatic analysis system for the mobile network base station traffic prediction and resource scheduling, and predicts the growth trend of the base station communication in a two-stage manner. In the first stage, the base mobile communication forecast result is used to estimate the base station news. In the second stage, we will fine-tune the forecasting trend of the base station's traffic by the characteristics of individual base stations and the space of the space. The base station traffic forecast results generated by the regional traffic growth speed weight threshold can be more reasonable and accurate. Forecasting; the traffic forecasting trend of individual base stations through the intelligent resource scheduling mechanism, in addition to predicting the base station that lacks capacity due to the growth of the traffic in the future, it can also screen out the base stations with redundant resources to provide scheduling in the future. The automatic analysis of the resource adjustment of the base station generates a new combination of platform combination and single-schedule scheduling, which achieves the optimization of the capacity planning and deployment of the base station and effectively improves the equipment utilization.

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 faster, and in the past it was only loaded for specific mobile network devices. Monitoring, if the monitoring threshold is reached, then purchase expansion. However, the limited wireless transmission resources of mobile network devices are not the same as the wired network, as long as the load reaches the specific threshold of the network device. Take the mobile network base station as an example, and build a large number of stations. Increasing the spatial density is not only costly, but also causes mutual interference between stations. Blindly amplifying mobile network equipment cannot effectively solve the problem of network capacity bottleneck.

In addition, the most similar patent in the current forecasting technology is the Chinese patent number: 201010300133, which is mainly used for the mobile network traffic to predict the growth trend of the SAS commercial software system for the prediction. The SAS system is a commercial software system. The cost is high and it is not easy to use and build. In addition, its traffic forecasting only uses the historical traffic volume of individual base stations, ignoring the influence of the number of users in the overall area and the growth trend of single-person usage, and does not consider the relationship between regional traffic growth trend and individual base stations. It will often encounter frequent fluctuations in the historical traffic of the base station, and the prediction results will be unreasonable. It is impossible to provide telecom operators as a reference for future capacity planning or resource allocation.

It can be seen that there is still much room for improvement in the above patents. For the construction of mobile networks and the need for maintenance, it is not a good design. It is urgent to consider the purpose and practicality of forecasting and quickly correct its shortcomings.

In view of the shortcomings derived from the above-mentioned conventional methods, the inventors of the present invention have improved and innovated, and after years of research, they have successfully developed and completed the automated analysis system and method for the mobile station base station traffic prediction and resource scheduling.

The purpose of the present invention is to provide a telecom operator a mobile network base station traffic growth prediction and resource scheduling automatic analysis system, which has automatic collection of mobile network historical traffic and device capacity configuration information, and translates into different according to different data dimensions. Input parameters and reference factors, combined with mathematical linear regression model to predict traffic trends and analyze the use of resources, through the human-machine interface to present the base station traffic forecast trend, forecast cycle capacity alarm and capacity resource adjustment base station new platform With monomer Scheduling matching combination; predictive periodic capacity alarm and capacity resource adjustment base station matching combination can be obtained through intelligent resource scheduling mechanism, and the obtained result is an important reference basis for mobile network base station capacity planning, which can assist telecom operators to deploy early. Plan, adjust and understand network load, improve network resource usage and optimize network performance.

The invention relates to a mobile network base station traffic prediction and resource scheduling automation system, which firstly uses the base station component and the interface data to interface with the external system data, and collects the input parameter data source of the base station traffic prediction, including the base station. Information such as traffic, configuration and regional traffic prediction, and then combined into data input parameters used by traffic prediction through different combinations of data; time dimension data, such as base station daily traffic volume of observation interval, daily Busy time traffic volume, daily average traffic volume, global traffic forecasting data growth rate; spatial dimension data, such as the daily traffic volume of the region and the regional classification of the base station capacity configuration; management dimension data, For example, the transmission capacity of the base station, the configuration status of the spectrum resources, and the proportion of the terminal data in the whole area, the input parameters are combined into input parameters through multi-dimensional translation.

From the input parameters to the prediction results, the phase prediction process is generated. The first phase is the traffic rough estimation phase. The historical traffic during the base station observation period is substituted into the linear regression model to obtain the traffic prediction reference value and the base station traffic slope. If the traffic slope is positive, the future traffic of each forecast cycle should consider the traffic growth trend of the whole region and the busy time impact of the base station. Each prediction cycle can calculate the rough estimate result of the busy time of the traffic; The traffic slope is negative, and the prediction result only considers the busy time prediction reference value.

The second stage is the fine-tuning phase of the traffic, which will fine-tune the traffic prediction rough estimate according to the base station's traffic characteristics. The growth rate factor of the base station is the fine-tuning factor obtained according to the base station's traffic characteristics, and express the base station's growth speed. Compared with the growth rate of the region's traffic, the base station grows. The higher the speed factor, the faster the growth rate of the base station's traffic is greater than the speed of the traffic in the region; the traffic fine-tuning result can be obtained by multiplying the traffic prediction rough estimate by the base station growth rate factor.

It can be found from the result of fine-tuning that if the historical information fluctuation of the base station is too large, the base station will grow too fast. This is an unreasonable prediction result, so it needs to be adjusted for the growth rate factor of the base station; The base station growth rate factor is the benchmark. After sorting by size, the base station growth rate factor upper and lower thresholds are set. This threshold is adjusted for the base station growth rate factor; the adjusted growth factor will reduce the base station traffic due to frequent fluctuations. The growth rate is too large. By multiplying the adjusted base station growth rate factor, fine-tuning the rough estimate results in a more reasonable and accurate forecast.

The third stage is the resource scheduling stage. Through the intelligent base station resource scheduling mechanism, the list of base stations with adjustable capacity alarm and capacity resources can be automatically obtained. Then, through the capacity configuration information (base station model, carrier open number, etc.), The combination of the newly added platform of the base station and the single scheduling can be quickly obtained, and the screening result can be adjusted and allocated according to the needs of the system user.

The results of traffic prediction and capacity alarms will be presented graphically in the human-machine interface query; the smart base station resource scheduling mechanism will also automatically update the base station capacity utilization and adjustable capacity during the forecast period. Provide an important reference for telecom operators to accurately build and cost control.

System automation is performed in a centrally controlled manner, managing automated start-up of phased processes, performing status monitoring and recording of error messages.

10‧‧‧System users

11‧‧‧External functions

12‧‧‧Mobile Network

20‧‧‧External data interface module

21‧‧‧Base station component and interface data acquisition module

22‧‧‧Multidimensional data translation module

23‧‧‧ Data Storage Module

24‧‧‧Predictive Computing Module

25‧‧‧ Traffic Analysis Module

26‧‧‧Smart Resource Dispatching Module

27‧‧‧Central Control Module

28‧‧‧Human Machine Interface

100‧‧‧ Input parameter source interface and combination

101‧‧‧Base station traffic coarse measurement

102‧‧‧Base station traffic forecasting fine-tuning preparation

103‧‧‧Base station traffic forecasting fine-tuning

104‧‧‧Base station capacity warning

105‧‧‧Base station resource scheduling

10501~10514‧‧‧Step process

1 is a system architecture diagram of the present invention; FIG. 2 is a flow chart of the present invention; FIG. 3 is a flow chart of smart capacity resource adjustment according to the present invention.

Referring to FIG. 1 , the mobile network base station traffic prediction and resource scheduling automatic analysis system of the present invention has three participants and nine functional modules, and the participants are system users 10, external systems 11 and mobile networks. The function module part is an external data interface module 20, a data interface and component data acquisition module 21, a multi-dimensional data translation module 22, a data storage module 23, a prediction operation module 24, and a message module. The analysis module 25, the intelligent area resource scheduling module 27, the central control module 28 and the human machine interface module 29; then the flow chart of Fig. 2 illustrates the automatic operation mode of the present invention in each functional module:

Step 1. The source of the input parameter is interfaced with the combination 100. The input parameter of the base station traffic prediction is derived from the external system 11 and the mobile network 12, and the external system 11 is the provider of the total regional terminal classification data volume and the predicted data volume. The mobile network 12 is a provider of base station traffic and configuration information; the system function uses the external data interface module 20 to retrieve the data provided by the external system 11, and the external system 11 provides the data mode in a standard file format. The external data interface module 20 interfaces the external system 11 according to the data format and interface of the provider, and the interface is interfaced with a standard network service data interface; the mobile network 12 communicates through the network service interface. And the component collects the base station communication and configuration parameter data, and provides the data mode in the standard file format. The base station interface and component data acquisition module 21 will operate the network according to the data format and interface of the provider. The data is introduced, and the interface can be interfaced by a standard file transfer method.

The input parameters of the base station traffic prediction need to be composed of different dimensional classification conversion source data, and the multi-dimensional data translation module 22 will design and be responsible for data translation of different dimensions, multi-dimensional data. The translation module 22 is responsible for translating data of three dimensions: time dimension, space dimension and dimension dimension. Time dimension translation integrates the input source data by time interval, such as the traffic volume during the observation period, and the daily interval through the time interval. Time-divisional time-differentiated data translation, which can translate the input parameter data of the base station's daily traffic volume, busy hour traffic volume and average traffic volume, externally communicated regional traffic prediction data volume data, and time interval Predicting the time-division data translation of the season, can translate the regional traffic forecasting rate of the regional traffic forecast data growth rate for each season; spatial dimension translation will integrate the input source data into the spatial region categories, such as the base during the observation period. The traffic information and configuration characteristics of the station can be translated into the spatial dimension data through the county area of the space area category base station, and the daily traffic volume of the county area can be translated. The configuration characteristics data can also be translated into the county. Base station configuration characteristics data of the city area; the dimension transfer will load the input source data into the user-defined dimension category. Statistical integration, such as the externally-connected terminal classification data volume, through the dimension classification of the dimension class to do the dimensional data translation of the dimension, the translation ratio of the terminal classification data can be translated, and the configuration characteristics of the base station can also be The dimension classification configuration of the maintenance category is used to translate the dimensional data of the dimension, and the configuration data of the base station with the transmission capacity of the base station and the configuration status of the spectrum resource is translated. The data translated by the multi-dimensional data translation module 22 is the base station information. The predicted input parameters are stored in the data storage module 23 after the input parameter combination is completed, and the data storage method can be stored in the memory or the database.

Step 2: Complete the base station traffic coarse test 101, start the base station traffic rough test, and first take out the daily traffic volume of the base station and the daily traffic volume of the base station in the area during the observation period from the data storage module 23. Sum, the input prediction computing module 24, the prediction computing module 24 performs mathematical operations on the linear prediction component, and the output result is the base station traffic prediction reference value and the traffic slope; the traffic analysis module 25 according to the prediction computing module 24 and the data storage module 23 data, rough estimate the base station traffic trend and store it in the data storage module 23, the data storage method can be stored in the memory or the database.

Step 3: Base station traffic forecasting preparations before fine-tuning 102. After completing the base station traffic coarse measurement, in order to make the traffic prediction trend closer to the actual traffic trend, special consideration is given to the individual base station traffic growth characteristics to fine-tune, but Before the fine-tuning, the regional traffic characteristics of the base station are first correlated with the influence factors of the base station's traffic characteristics, as a preparation before the fine-tuning; before the fine-tuning preparation, the base station traffic is first taken out from the data storage module 23. The slope and the traffic slope of the region, the input traffic analysis module 25 calculates the growth rate factor of the individual base stations; the growth rate factor of all the base stations classified by the region will be in the traffic analysis module 25 according to the growth rate factor of the base station. Sort the size, then set the screening threshold, and filter out the high and low thresholds of the growth rate factor of the base station. The setting method of the screening threshold can be set by the system parameter or input by the system user through the human interface module 29, and the threshold is filtered to Threshold _{toprank %} Said, the selected base station growth rate factor is the former Thre of all base stations under the jurisdiction of the region. The base station growth rate factor of shold _{toprank %} and post-Threshold _{toprank %} , the selected two base station growth rate factors are the upper and lower limits of the growth rate factor of the base station under the regional jurisdiction, if the growth rate factor of the base station of the regional base station is greater than The base station growth rate factor of the former Threshold _{toprank %} is automatically adjusted to the upper limit value. Similarly, if the growth rate factor of the base station under the regional base is less than the lower limit value of the base station growth rate factor of the subsequent Threshold _{toprank %} , It will be automatically adjusted to the lower limit value. The adjusted base station growth rate factor will be stored in the data storage module 23, and the data storage method can be stored in the memory or the database.

Step 4: Complete the base station traffic forecasting fine-tuning 103, start the base station traffic forecasting fine-tuning, and first obtain the base station traffic slope, the base station prediction rough estimation result and the growth speed factor from the data storage module 23, and input the traffic analysis module. Group 25; the traffic analysis module 25 uses the base station traffic slope to determine, if the base station traffic is growing, the base station traffic busy time is roughly estimated as the base station traffic prediction rough estimation result, and each prediction period is The busy time rough estimate result is multiplied by the adjusted base station growth rate factor. It can complete the fine-tuning of the traffic information of the base station's traffic growth; if the base station's traffic is negative, the base station's busy time prediction reference value and the adjusted base station growth rate factor are taken out, and the prediction computing module 24 is input. The busy baseline prediction value is the prediction reference. The adjusted base station growth rate factor is the traffic slope. The traffic prediction fine-tuning result for each prediction period is calculated by linear regression. If the negative growth traffic prediction fine-tuning result is negative. The base station will be stored in the data storage module 23. The data storage method will be stored in the database. The storage content can be provided by the system user through the human interface module 29. .

Step 5: Complete the base station capacity warning 104, start the base station capacity warning, and first take out the base station traffic prediction fine-tuning result and the base station capacity configuration data in the input parameter from the data storage module 23, and the base station capacity configuration data. There is information about the area, transmission capacity and spectrum resource configuration status of the base station. The extracted information will be input into the intelligent resource scheduling module 27. The intelligent resource scheduling module 27 will take the base station transmission capacity of the base station capacity configuration data. The capacity alarm line is set according to the capacity alarm threshold. The capacity alarm threshold setting can be set by the system parameter or by the system user through the human interface module 29. The capacity alarm threshold is expressed by the Proportion _{capacityalarm} , and the capacity alarm line is Proportion. The transmission capacity of the _{capacityalarm} multiple; after obtaining the base station capacity alarm line, the fine-tuning result of the base station traffic prediction is compared according to its prediction period, and if the base station capacity alarm line is exceeded, the prediction period base station communication is indicated. Growth capacity alarm, intelligent regional resource scheduling module 27 stores the result of the base station traffic prediction fine-tuning result in each prediction period and the capacity alarm line in the data storage module 23. The data storage method is stored in the database, and the stored content can be provided to the system user. The machine interface module 29 queries.

Step 6. Complete the base station resource dispatching 105. The capacity of the base station is limited by the planning and configuration of the base station capacity of the mobile network. The capacity of the base station is newly built to expand the capacity resources. The neighboring avoidance effect of the station is not easy to achieve; at present, the base station's traffic capacity expansion is based on the traffic channel component (CE, Channel Element), and the software expansion (License) mode is used to complete the capacity expansion. This expansion mode only requires software authentication setting expansion. There is no need to add new equipment. In the future, the evolution of the capacity expansion mode of the base station will avoid the neighboring avoidance effect and will be expanded towards software authentication.

The intelligent resource scheduling module 26 first extracts the prediction result of the base station traffic prediction and the transmission capacity information of the base station from the data storage module 23, and inputs the smart resource scheduling module 27, and the smart resource scheduling module 27 The results of the smart resource scheduling mechanism are stored in the data storage module 23. The data storage method is stored in the database, and the stored content can be provided by the system user through the human interface module 29.

The intelligent resource scheduling mechanism is shown in Figure 3. The intelligent resource scheduling operation process can periodically filter the base station with adjustable capacity according to the traffic prediction result and the capacity configuration parameter, and also filter the base station for the executable resource scheduling matching. First, after the smart resource scheduling 10501 program starts, the smart resource scheduling module 27 calculates the base station capacity resource usage rate 10502.

After the base station capacity resource utilization is calculated, it will enter the minimum adjustable capacity ratio screening 10503 as the screening threshold for capacity resource utilization; if the capacity resource utilization ratio is greater than or equal to the threshold, it can screen out the base with sufficient capacity and adjustable resources. The combination 10504, on the other hand, if it is smaller than the threshold, the base station combination 10506 with insufficient capacity and unable to provide the adjustment resource can be selected; the obtained capacity is sufficient and the result of the adjustment base station 10504 can be provided, and the base station adjustable capacity 10505 can be quickly calculated.

The base station 10506 with insufficient capacity and unable to provide the adjustment resource will perform the capacity alarm base station screening 10507 (same step 5). If the base station simultaneously generates the capacity alarm, it indicates that the base station urgently needs to perform resource scheduling in the prediction period, and vice versa. A capacity alarm occurs, indicating that the base station cannot provide the adjustment status only for the capacity, and the capacity resources held can still be used continuously. At this time, the adjustment terminal 10508 is directly entered; The base station with the capacity alarm in the forecast period will be filtered out, and the selected station will perform the unadjusted base station screening 10509. The unadjustable base station screening, that is, the base station that screens the capacity alarm, the traffic of the prediction period is fully loaded and The available spectrum resources have been exhausted. If both are used, the screening result will be listed as the new station combination 10510, indicating that the base stations in the combination need to increase the base station capacity by adding the new station mode. The screening result is a base station 10511 with insufficient capacity to be adjusted.

The base station combination with sufficient capacity and adjustable capacity and the list of base stations that generate capacity alarms and urgently need capacity resources are filtered out, and the matching information of the single scheduling is generated through the capacity configuration information of the associated base station, and the capacity configuration is performed. Correlation screening, the generated capacity alarm base station and the adjustable capacity base station single scheduling matching combination 10513, the new station combination 10510 and the single scheduling matching combination 10531 can be provided to the system user reference, the system user can be based on the capacity resource Plan the requirements, and select the base station combination 10514 that meets the conditions to perform resource scheduling.

As shown in the functional module architecture of FIG. 1, the central control module 28 has the automatic start-up mechanism of each step described in the process steps of FIG. 2. After the step is started, the central control module 28 monitors the automatic execution status of each step, and if the step is successfully executed. The automatic switching step is performed to execute the next step, and the step status and execution time are stored in the data storage module 23; if an error occurs during the step execution, the error occurrence alarm message is stored in the data storage module 23; the central control module The data storage method of 23 will be stored in the execution log file, and the storage content can provide a reason for the system developer and the system user to encounter data errors and execution failures.

The mobile network base station traffic prediction and resource scheduling automatic analysis system provided by the invention can be clearly found when compared with other conventional technologies, and the existing voice traffic prediction method is applied, and the historical information of the individual base stations is used. The forecast of traffic volume, the forecast results show frequent divergence or unreasonable growth trend, and the prediction results are limited to the traffic analysis of individual base stations; The patented function module, process steps and intelligent resource scheduling mechanism can not only avoid the divergence or unreasonable results of the base station traffic prediction trend, but also the traffic prediction result combined with the capacity configuration to estimate the mobile network base station. The capacity resource can be adjusted and the combination of the newly added base station with insufficient capacity and the base station single unit scheduling match, which provides an important reference basis for system users to plan and adjust the capacity in the future.

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.

10‧‧‧System users

11‧‧‧External functions

12‧‧‧Mobile Network

20‧‧‧External data interface module

21‧‧‧Base station component and interface data acquisition module

22‧‧‧Multidimensional data translation module

23‧‧‧ Data Storage Module

24‧‧‧Predictive Computing Module

25‧‧‧ Traffic Analysis Module

26‧‧‧Smart Resource Dispatching Module

27‧‧‧Central Control Module

28‧‧‧Human Machine Interface

Claims (2)

An automatic analysis method for mobile network base station traffic prediction and resource scheduling, the steps of which include: a. input parameter source data interface and combination: translating external interface data into a regional terminal classification through a multi-dimensional data translation module The data ratio and the growth rate of the regional traffic prediction data, the translation of the base station components and interface interface data, the daily traffic volume of the base station, the daily busy hour traffic volume, the daily average traffic volume, and the area The total daily traffic volume of the base station and the base station capacity configuration information; b. The base station traffic coarse measurement: predictive base station traffic prediction reference value and base station traffic using linear prediction operation using predictive computing module Slope and regional classification of traffic slope; c. Base station traffic prediction fine-tuning preparation: The base station's traffic volume is correlated with the regional base station traffic usage traffic analysis module to obtain the base station growth rate factor. d. Base station traffic forecasting fine-tuning: The intelligent resource scheduling module judges the positive and negative growth of the traffic for the base station traffic slope. If it is judged to be positive growth, the fine-tuning result is the base. The traffic rough estimation result of each prediction period is multiplied by the growth rate factor; if it is negative growth, the base station traffic prediction reference value is the prediction reference, the growth speed factor is the prediction slope, and each prediction period is calculated by linear regression. Traffic prediction and fine-tuning results; e. Base station capacity warning: The intelligent resource scheduling module compares the base station traffic prediction trend and the base station capacity configuration information to determine whether there is a capacity alarm in the future; f. Base station resource scheduling: The intelligent resource scheduling module combines the base station's future traffic forecasting trend and the base station capacity configuration information to automatically filter out the base stations with redundant resources for scheduling and the schedulable capacity. The method for automatically analyzing the mobile network base station traffic prediction and resource scheduling according to claim 1, wherein the method further comprises: monitoring the automatic execution status of each step through a central control module, and executing The function will automatically switch to the next step until the end of the process. If the execution fails, the execution error will be recorded in the error log for subsequent trace debugging.