WO2023170837A1

WO2023170837A1 - Communication bandwidth calculation device, communication bandwidth calculation method, and program

Info

Publication number: WO2023170837A1
Application number: PCT/JP2022/010387
Authority: WO
Inventors: 正次高野; 恵竹下
Original assignee: 日本電信電話株式会社
Priority date: 2022-03-09
Filing date: 2022-03-09
Publication date: 2023-09-14

Abstract

This communication bandwidth calculation device, which calculates the required bandwidth of communication facilities, is provided with: an information acquisition unit that acquires traffic information for each communication facility, and attribute statistic traffic information and corresponding attribute statistic attribute information which are statistically processed on the basis of traffic information and contract user information for each communication terminal; a prediction calculation unit that calculates macro traffic growth rate prediction information by analysis using the attribute statistic traffic information and the attribute statistic attribute information; and a required bandwidth calculation unit that calculates the required bandwidth of each communication facility on the basis of the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction factor for each communication facility.

Description

Communication band calculation device, communication band calculation method, and program

The present invention relates to a technique for calculating the required bandwidth of communication equipment in a communication network.

Traditionally, there have been requirements from users for communication services provided via communication networks, regardless of whether the communication is via a fixed line (wired) or wireless communication via a mobile device. QoS (Quality of Service) or QoE (Quality of Experience) can be defined as the quality of communication service. Communication service providers design, operate, and manage communication networks that achieve such service quality.

In order to achieve the above objectives, in communication networks, traffic volume is regularly measured and, at the same time, the traffic characteristics of the communication services provided are analyzed and evaluated to gain knowledge of the traffic characteristics. Utilizing the knowledge obtained, we can predict the amount of communication traffic in the future and achieve the quality of communication services required by users under conditions where the predicted traffic amount is a burden on the communication network. Therefore, there is a need for a technology to calculate the correct amount of communication resources to ensure the economic efficiency of communication services.

Conventional technology has been developed to predict the amount of communication traffic in the future, for communication networks that provide fixed-line telephone services and multiplexed various communication services, and for communication networks that provide secure communication services that connect LANs of corporate bases. There are many techniques for calculating the amount of communication equipment. Patent Document 1 is one of these techniques.

Japanese Patent Application Publication No. 2020-150524

Regardless of whether it is a fixed line (wired) communication or a mobile wireless communication, communication services such as Internet connection services, video distribution services, VPN services, games, IP telephones, videophones, SNS, etc. Diversification is progressing rapidly. When using these communication services, there is a large difference in the amount of data transferred per unit time.

At the same time, the communication services used by individual users are largely biased depending on their preferences. As a result, there are extremely large differences in the amount of transferred data consumed by individual users and the speed at which it increases, making it increasingly difficult to predict future traffic amounts.

On the other hand, with the progress of digitalization of social life, there is a strong demand for the protection of personal information, and the use of personal information managed by telecommunications carriers for the design, operation, and management of communication traffic is strictly restricted.

The present invention has been made in view of the above points, and is intended to predict future traffic volume and accurately calculate the communication band required to achieve the quality of communication services while protecting personal information. The purpose is to provide the following technology.

According to the disclosed technology, there is provided a communication band calculation device that calculates the required band of communication equipment of a communication network,
an information acquisition unit that acquires attribute statistical traffic information and corresponding attribute statistical attribute information statistically processed based on traffic information for each communication terminal and contracted user information, and traffic information for each communication facility;
a prediction calculation unit that calculates macro traffic growth rate prediction information by analysis using the attribute statistical traffic information and the attribute statistical attribute information;
a required bandwidth calculation unit that calculates a required bandwidth for each communication facility based on the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction coefficient for each communication facility;
A communication band calculation device is provided.

According to the disclosed technology, it is possible to predict future traffic volume and accurately calculate the communication band required to achieve the quality of communication services while protecting personal information.

1 is an overall block diagram including the configuration of a communication band calculation device according to an embodiment of the present invention. FIG. FIG. 2 is a block diagram showing the internal configuration of an arithmetic processing unit according to an embodiment of the present invention. FIG. 2 is a flow diagram (Part 1) showing processing of the arithmetic processing unit. FIG. 3 is a flowchart (part 2) showing the processing of the arithmetic processing unit. FIG. 3 is a flow diagram (part 3) showing the processing of the arithmetic processing unit. FIG. 7 is a flow diagram showing another configuration of the traffic prediction calculation unit.

Hereinafter, an embodiment of the present invention (this embodiment) will be described with reference to the drawings. The embodiments described below are merely examples, and embodiments to which the present invention is applied are not limited to the following embodiments.

(Summary of embodiment)
This embodiment targets communication equipment in a communication network where the amount of communication traffic increases (decreases) while changing in a complex manner, and aims to economically provide the communication service quality required by the users who use it. Next, we will explain the technology for predicting and calculating the amount of bandwidth equipment that will be required at a future design target period.

The above calculation process is executed by the communication band calculation device 10, which will be described later. More specifically, the communication bandwidth calculation device 10 calculates the load traffic on the entire communication network to be designed by predicting the amount of transferred data for each attribute at a future design target period based only on information made statistical based on the attributes. To accurately calculate the communication bandwidth required to achieve the quality of communication service.

In this embodiment, a situation is assumed in which calculation and evaluation of the amount of bandwidth equipment that will be required in the future is continuously performed as a communication carrier manages and operates communication network equipment. Calculate the bandwidth required to ensure the quality of communication services at a desired point in the future, using a certain period as a cycle, estimate the extent of future bottlenecks and strains on existing bandwidth equipment, and calculate the difference between the two. Plan and execute construction work such as equipment expansion. Hereinafter, this embodiment will be described in detail with reference to the drawings.

(Overall system configuration)
FIG. 1 is an overall block diagram showing a system configuration including a communication band calculation device 10 according to the present embodiment. In this embodiment, a communication network 20 that provides a data communication service will be described as an example of a communication network for which a communication band is to be calculated. In this embodiment, it is assumed that data communication is performed in the communication network 20 using the IP protocol. However, data communication using the IP protocol is just one example, and the technology according to the present invention is applicable regardless of the type of protocol.

As shown in FIG. 1, the communication network 20 is a facility whose purpose is to provide data communication services to communication terminals 41, 42, 43, and 44.

PC terminals

45, 46, 47, and 48 are connected to the communication terminals 41, 42, 43, and 44, respectively. The data information transmitted and received by the communication terminal is transferred to the node 21 via the

access nodes

23 and 24 and the band equipment 31 and 32, the node 22 and the band equipment 30, and then to the desired destination in the communication network 20. The information is sequentially transferred to the destination communication terminal. Data communication services are thereby achieved.

Note that the node is, for example, a router or a switch. Further, the band equipment may be called a transmission path, a communication path, a line, etc. Both nodes and band equipment are examples of communication equipment.

The communication band calculation device 10 is composed of an information processing device using a computer. The communication bandwidth calculation device 10 periodically or timely acquires network equipment configuration information 51 including information regarding communication equipment such as node information, line bandwidth information, and topology information regarding the communication network 20 from the operation system 61.

The equipment unit traffic information 52 shown in FIG. 1 will be explained. For example, the equipment unit traffic information 52 related to the band equipment 30 includes measurement data obtained by measuring the amount of traffic flowing out/inflowing from the band equipment 30 at regular time intervals at the

nodes

21 and 22. The operation system 61 maintains equipment unit traffic information 52 for all communication equipment that is subject to operation, management, and design.

The per-terminal traffic information 53 shown in FIG. 1 will be explained. For example, the terminal unit traffic information 53 related to the communication terminal 41 refers to communication data transmitted and received by the communication terminal 41 within a specified period (such as a billing period) specified in a data communication service, as identified by the communication terminal ID of the communication terminal 41. This is information calculated by counting the amount. The operation system 61 maintains terminal-based traffic information 53 for all communication terminals that provide communication services.

The communication band calculation device 10 acquires network equipment configuration information 51 and equipment unit traffic information 52 from the operation system 61 periodically or at a timely manner.

Telecommunications carriers manage information such as name, gender, date of birth (age), address, contract plan, communication terminal ID linked to the contract plan, and telephone number for users who contract data communication services. are doing. This management information is called user unit information 54. These management information items will be referred to as attribute items, and the specific contents will be referred to as attribute values. User unit information 54 regarding all users who contract for data communication services is held in the user/contract management system 62.

Additionally, the user/contract management system 62 obtains the per-terminal traffic information 53 from the operation system 61 on a regular or timely basis.

By comparing the communication terminal ID included in the per-terminal traffic information 53, for each communication terminal, the per-user information 54 such as the contracted user and the contracted plan, etc., and the user unit information 54 such as the contracted user and the contracted plan, and the use of the corresponding communication terminal within a specified period (billing period, etc.) It is possible to link terminal unit traffic information 53 such as the amount of communication data transmitted and received to the terminal.

The user unit information 54 includes personal information, and its use is extremely strictly restricted by the Personal Information Protection Act. Therefore, it is rational for telecommunications carriers to process information used in the design, management, and operation of telecommunications network equipment so that it does not include personal information. Here, processing information that includes personal information into information that does not include personal information (non-personal information) is called statisticization, and such processing processing is called statistical processing.

In this embodiment, as will be described in detail later, the term attribute statisticization is used to clearly indicate the state in which the attributes included in the user unit information 54 have become non-personal information through statistical processing.

The statistical processing device 63 shown in FIG. 1 acquires the terminal-based traffic information 53 and the user-based information 54 from the user/contract management system 62, and compares the two using the communication terminal ID to calculate user information and traffic information. Can relate to information.

The statistical processing device 63 can perform statistical processing using attribute information such as name, gender, date of birth (age), address, and contract plan included in the user unit information 54. For example, age can be calculated from the date of birth, and statistics can be made into user attributes for age groups such as 30s (30 to 39 years old). Addresses can be statisticized into user attributes at the prefecture level. In order to statisticize user unit information such as being in his 30s and living in Tokyo, information that specifies conditions for statisticization based on attribute items and their attribute values is called statisticization instruction information 47. The statistical instruction information 47 is acquired from the communication band calculation device 10.

Furthermore, the statistical processing device 63 determines, for example, that the user attribute is 30, based on the statistical instruction information 57, which is information that specifies the user attribute to be statisticized, for the traffic information related by matching using the communication terminal ID. Extract a set of communication terminals that correspond to the age group and reside in Tokyo, and use the history information of the amount of transferred data in the per-terminal traffic information 53 to identify communication terminals included in the section where the amount of transferred data is 5 GB or more and less than 10 GB. It is possible to make statistics as a number and output a histogram, for example.

In this way, based on the statistical instruction information 57, which is information that specifies the conditions for statistical analysis based on the attribute items included in the user attributes and their attribute values, statistical traffic information (non-personal information) is converted into attribute statistical traffic. It is called information 55.

Furthermore, information (non-personal information) that is statisticized based on user attributes other than traffic information, such as the number or share of users for each user attribute and the number of communication terminals, is referred to as attribute statisticized attribute information 56.

The communication band calculation device 10 acquires network equipment configuration information 51 and equipment unit traffic information 52 from the operation system 61 regularly or at any time. Furthermore, the communication band calculation device 10 acquires the attribute statistical traffic information 55 and the attribute statistical attribute information 56 from the statistical processing device 63 periodically or at any time. Further, the communication band calculation device 10 creates statistical instruction information 57, provides it to the statistical processing device 63 periodically or at any time, and uses it as information specifying conditions for statistical processing.

Since both the attribute statistical traffic information 55 and the attribute statistical attribute information 56 are already statistical information (non-personal information) that does not include personal information, the communication band calculation device 10 uses statistical information (non-personal information). (Personal Information) only and will not use personal information.

The communication bandwidth calculation device 10 predicts the amount of load traffic on the entire communication network to be designed by predicting the amount of transferred data for each attribute at a future design target period based only on the information made into statistics based on the acquired attributes. , it is possible to accurately calculate the communication band required to achieve the quality of communication service.

(Internal configuration of communication band calculation device 10)
Next, the internal configuration of the communication band calculation device 10 according to this embodiment will be described in detail.

The configuration of the communication band calculation device 10 shown in FIG. 1 shows an example of the hardware configuration when the communication band calculation device 10 is implemented by a computer. However, the communication bandwidth calculation device 10 may be a physical machine or a virtual machine, and when the communication bandwidth calculation device 10 is implemented as a virtual machine, the hardware configuration shown in FIG. 1 becomes a virtual hardware configuration.

As shown in FIG. 1, the communication band calculation device 10 includes a communication interface section 11 (hereinafter referred to as communication I/F section 11), an operation input section 12, a screen display section 13, and an information database section as main components. 14 (hereinafter referred to as the information DB section 14), a storage section 15, and an arithmetic processing section 16, which are connected via an internal communication bus and can mutually send and receive information.

The communication I/F section 11 is composed of a dedicated data communication circuit, and has a function of mutually communicating with external devices such as the operation system 61.

The operation input unit 12 consists of an operation input device such as a keyboard and a mouse, and has a function of detecting input operations from an operator and outputting them to the arithmetic processing unit 16.

The screen display unit 13 is a screen display device such as a display, and has a function of displaying various information such as operation menus and calculation results on the screen in response to instructions from the arithmetic processing unit 16.

The information DB unit 14 is comprised of a storage device such as a hard disk or a memory, and has a function of storing various data used in the required bandwidth calculation process in the arithmetic processing unit 16.

The storage unit 15 is composed of a storage device such as a hard disk or a memory, and has a function of storing various programs and data used in the required bandwidth calculation process in the arithmetic processing unit 16.

The arithmetic processing unit 16 includes a microprocessor such as a CPU (Central Processing Unit) and its peripheral circuits, and reads a program from the storage unit 15 and executes the program to perform operations from the information DB 14 or the operation input unit 12. Through this, network equipment configuration information 51, equipment unit traffic information 52, attribute statistical traffic information 55, attribute statistical attribute information 56, etc. required for calculation processing are acquired periodically or in a timely manner, and the information is statisticized by attributes. By predicting the amount of transferred data for each attribute in the future design target period based on the data, it is possible to predict the amount of load traffic that will be applied to the entire communication network to be designed, and to estimate the communication bandwidth required to achieve the quality of communication service. is calculated with high precision, and the calculation result is outputted to the information DB section 14 or the like.

A program that realizes the processing in the communication band calculation device 10 is provided, for example, on a recording medium such as a CD-ROM or a memory card. The program read from the recording medium is stored, for example, in the storage section 15, read out from the arithmetic processing section 16, and executed. Note that the program may be downloaded from a server or the like via a network.

(About various information)
The network equipment configuration information 51 shown in FIG. Bandwidth information B_{j} of any interface j of the communication equipment that is used is included. Further, the network equipment configuration information 51 includes information on connection configuration relationships between arbitrary interfaces of communication equipment managed and operated within the communication network 20.

Furthermore, the network equipment configuration information 51 includes connection configuration relationships between arbitrary interfaces of communication equipment operated in the communication network 20, not only at present but also from past equipment construction history to planned future equipment construction plans. information shall also be included. For simplicity, the network equipment configuration information 51 will be written as {BD}.

The equipment unit traffic information 52 includes measured traffic that is continuously measured over a long period of time at a predetermined measurement interval regarding the amount of traffic flowing in/out of any interface j of communication equipment operated in the communication network 20. Contains time series data of amount. Let Y_{j}(t) be the measured traffic amount of the interface j during the measurement period t, and define the set of time-series data as {Y_{j}(t),t∈T}. This will be referred to as time series data of the measured traffic amount, or simply as the measured traffic amount. In principle, the measured traffic volume continues to be measured for all communication equipment and their interfaces. For simplicity, the equipment unit traffic information 52 is written as {DT}.

The per-terminal traffic information 53 will be explained. As described above, for any user who subscribes to a data communication service of a communication carrier, correspondence of communication terminals tied to the contents of the contract is managed. The terminal-based traffic information 53 is communication data amount data accumulated for each communication terminal over a certain period of time (for example, monthly), and includes past history information. For simplicity, the terminal unit traffic information 53 is written as {TT}.

The user unit information 54 will be explained. As mentioned above, for the purpose of billing and providing appropriate services, telecommunications carriers provide data communication service contract users with their name, gender, date of birth (age), address, contract plan, and contract plan information. It manages information such as linked communication terminal IDs and telephone numbers, and this management information is referred to as user unit information 54.

The attribute statistical traffic information 55 and the attribute statistical attribute information 56 will be explained. As described above, by comparing the terminal unit traffic information 53 and the user unit information 54 using the communication terminal ID commonly included in both, the user information and the traffic information are related, and then the above-mentioned The traffic information that has been statisticized based on the statisticization instruction information 57 is referred to as attribute statisticization traffic information 55. Statistical information other than traffic information, such as the number of users, share, and number of communication terminals for each user attribute, is referred to as attribute statistical attribute information 56.

The statistical instruction information 57 will be explained. As described above, the statistical instruction information 57 is used as a condition for statistical processing in order to perform statistical processing for the purpose of processing user unit information 54 and terminal unit traffic information 53, which include personal information, into non-personal information. This information includes specified attribute items, attribute values, specific set operations, and statistical processing. The statisticalization instruction information 57 is set (created) within the communication band calculation device 10, acquired periodically or at any time by the statistical processing device 63, and applied to statistical processing.

(Configuration of arithmetic processing unit 16)
Next, with reference to FIG. 2, the internal configuration of the arithmetic processing section 16 according to this embodiment will be described in detail. FIG. 2 is a block diagram showing each processing unit for calculating a communication band in the arithmetic processing unit 16.

The internal configuration of the arithmetic processing unit 16 shown in FIG. 2 corresponds to a functional configuration realized by the arithmetic processing unit 16 when the arithmetic processing unit 16 executes a program. The internal configuration of the arithmetic processing unit 16 shown in FIG. 2 may be interpreted as the functional configuration of the communication band calculation device 10.

As shown in FIG. 2, the calculation processing unit 16 includes an information acquisition unit 16A, a prediction design unit 16B, a prediction calculation unit 16C, and a required bandwidth calculation unit 16D as main processing units.

The information acquisition unit 16A periodically or periodically obtains network equipment configuration information 51 and equipment unit traffic information 52, which are information necessary for communication band calculation, from the operation system 61 regarding the communication equipment of the communication network 20 that is the design target. Obtain at any time. Further, the information acquisition unit 16A acquires the attribute statistical traffic information 55 and the attribute statistical attribute information 56 from the statistical processing device 63 periodically or at any time.

As described above, in the statistical processing device 63, the terminal unit traffic information 53 and the user unit information 54 are input, and the attribute traffic information 55 and the attribute statistical attribute information 56 are created by statistical processing. The prediction design unit 16B creates statistical instruction information 57, which is the conditions or procedures for statistical processing for the attribute items and attribute values necessary for the statistical processing.

Since the statistical instruction information 57 needs to be an attribute item of the user whose description is managed in the user unit information 54 managed in the user/contract management system 62, the predictive design unit 16B Obtained from the information DB section 14. The attribute item may be stored in the information DB section 14 as information using the operation input section 12 or as information acquired through communication with the user/contract management system 62.

The prediction design unit 16B creates statistical instruction information 57, which is the condition or procedure for statistical processing, from the attribute items.

As mentioned above, the age of each contracted user can be calculated from the date of birth, integrated in 10-year increments, and statisticized into user attributes based on age. Addresses can be integrated by prefecture and compiled into user attributes for each prefecture. Furthermore, it is also possible to generate a histogram of the amount of transferred data for users with an AND condition (intersection set) of user attributes of age and prefecture. Since individuals cannot be identified from such statistical data, it is considered non-personal information.

Such statistical processing can generally be described by set operations on attribute items and their attribute values. The created statistical instruction information 57 is acquired by the statistical processing device 63 periodically or at any time.

The results of traffic prediction and share prediction can vary greatly depending on the instruction contents of the statistical instruction information 57, so it is expected that the accuracy of the prediction results will be high by reconsidering the results of various predictions as described below. Adjust attribute items and their attribute values to design optimal statistical instruction information. The statistical instruction information 57 created by the prediction design section 16B is stored in the information DB section 14.

The prediction calculation unit 16C acquires attribute statistical traffic information 55, attribute statistical attribute information 56, network equipment configuration information 41, and equipment unit traffic information 52 from the information acquisition unit 16A.

The prediction calculation unit 16C inputs this information and calculates attribute statistical traffic prediction information 71, attribute statistical share prediction information 72, and user average traffic prediction information related to future traffic prediction by regression prediction, which will be described in more detail later. 73, contract number prediction information 74 and macro traffic growth rate prediction information 75 are calculated.

Attribute statistical traffic prediction information 71, attribute statistical share prediction information 72, user average traffic prediction information 73, number of contracts prediction information 74, and macro traffic growth rate prediction information 75 are stored in the information DB unit 14.

The required bandwidth calculation unit 16D acquires network equipment configuration information 51 and equipment unit traffic information 52 from the information acquisition unit 16, and acquires macro traffic growth rate prediction information 75 from the prediction calculation unit 16C. , correction coefficient information 77 is acquired from the information DB section 14. Using this information, the required bandwidth calculation unit 16D calculates required bandwidth information 76, which is information on the required bandwidth for the design target period, for all target communication equipment.

(About the required bandwidth)
Here, the required bandwidth will be explained. The traffic data included in the equipment unit traffic information 52 is the average traffic flow rate bit/sec in each communication equipment at a measurement time granularity of 1 hour or 5 minutes, which is shorter than the measurement time granularity. At time granularity, there is always a moment when the actual traffic flow exceeds the measured traffic amount even within the same measurement time.

The transmission of IP packets flowing in a communication flow based on the IP protocol is not at a uniform speed, but has large deviations. This instantaneous unevenness of IP packets is called burstiness. The property that the actual traffic volume momentarily exceeds the measured traffic volume for the same measurement period due to the strong influence of the bursty nature of the flow of IP packets is called short-term traffic fluctuation.

In other words, in the future design target period, short-term fluctuations in traffic cannot be absorbed simply by having band equipment with the same equipment capacity as the measured traffic amount.

Therefore, in order to ensure the communication quality expected by users in communication services, a band that can reliably absorb short-term fluctuations in traffic is required. In this way, it is necessary to calculate an optimal band that is larger than the measured traffic amount, can absorb short-term fluctuations in traffic, and is not uneconomical. The optimal band that takes such short-term fluctuations into account will be referred to as the "required band."

The coefficient for converting the above-mentioned measured traffic amount into the required band will be referred to as a correction coefficient. The correction coefficient is a correction coefficient for filling the following gaps, and can be set for each communication facility.

First, the amount of traffic measured by communication equipment is a value averaged over the measurement time interval. On the other hand, in actual communication, there is burstiness at the IP packet level, and there are moments when the traffic becomes larger than the measured value. Because there is a scale gap between the time scale for measuring the amount of traffic in communication equipment and the time scale at which burstiness of IP packets occurs, depending on the burstiness of IP packets, the bandwidth equivalent to the amount of traffic to be measured may vary depending on the burstiness of IP packets. In this case, IP packets may be lost. Therefore, extra bandwidth is required to absorb the burstiness of IP packets and prevent loss of IP packets. The required bandwidth is the measured traffic amount plus the surplus bandwidth, and in that sense, correction is required between the measured traffic amount and the required bandwidth.

Equipment verification, such as measuring short-term fluctuations in IP packets using packet capture, etc., under conditions where actual (including simulated) communication traffic loads are applied to commonly used communication equipment, It becomes possible to calculate a correction value for calculating the required band for the measured traffic amount from the performance specifications of the communication device.

Second, the measurement time interval of the original data used to calculate the traffic volume as a macro (for example, the amount of personal transfer data is accumulated on a monthly basis) and the measurement time interval of the original data used to calculate the traffic volume as a macro, and the traffic volume measured by communication equipment. There is a large scale gap between time intervals (eg, 5 minutes). Furthermore, since communication service demand or traffic demand has large temporal fluctuations such as during busy times and off-peak times, it is important to consider the macro level of traffic (including its predicted value) and the load measured by communication equipment. It is necessary to correct the difference between the maximum traffic amount and the maximum traffic amount.

However, the demand for communication services and the busiest times are relatively stable as they are strongly linked to social life, so it is possible to calculate the correction value by verifying past data.

Third, since the macro traffic growth rate and the traffic growth rate of individual communication facilities generally do not match, it is also necessary to correct the gap between them. However, this correction can also be calculated by verifying past data.

The combination of the above-mentioned correction values for correcting the three gaps related to individual communication equipment is called a correction value. A collection of correction values for individual communication equipment for the entire communication equipment is called correction coefficient information 77. The correction coefficient information 77 is generated and managed by the information DB section 14.

Note that the correction values forming the correction coefficient information 77 are not limited to a combination of correction values for correcting the three gaps described above. A correction value for correcting gaps other than the three gaps described above may be used, or a correction value for correcting any one or any two of the three gaps may be used.

The required bandwidth calculation unit 16D obtains macro traffic growth rate information 75, equipment unit traffic information 52, and correction coefficient information 77.

Then, the required bandwidth calculation unit 16D calculates the macro traffic growth rate r(m) at the design target period m included in the macro traffic growth rate information 75 and the equipment unit The traffic amount Y_j(0) measured most recently for the communication equipment_j included in the traffic information 52 (for simplicity, the measurement time is set to 0) and the traffic volume Y_j(0) for the communication equipment_j included in the correction coefficient information 77. Using the correction coefficient K_j, the required band B_j(m) for the design target period m is calculated by the following formula.

B_j(m)= r(m) * Y_j(0) * K_j
The required bandwidth calculation unit 16D stores the required bandwidth information 77 in the information DB unit 14.

(Processing of prediction calculation unit)
Next, the processing of the prediction calculation unit 16C according to this embodiment will be described in detail with reference to FIGS. 3 to 5. 3 to 5 are flowcharts showing the processing of the prediction calculation unit 16C. In these flowcharts, it is also described in which functional unit within the prediction calculation unit 16C the processing of each step is executed.

First, with reference to FIG. 3, the creation process of attribute statistical traffic prediction information and attribute statistical share prediction information will be described. In S110, the prediction calculation control unit 16C1 acquires the attribute statistical traffic information 55 and the attribute statistical attribute information 56 from the information acquisition unit 16A. The attribute statistical traffic information 55 and the attribute statistical attribute information 56 are written as {ST} and {AT}, respectively.

In addition, in S120, the prediction calculation control unit 16C1 receives regression prediction calculation parameters including the timing to predict, the period to be used for regression, etc., which are necessary to execute the regression prediction calculation described later, from the prediction design unit 16B. get. The regression prediction calculation parameter is written as {R}.

Next, a regression prediction calculation algorithm used in this embodiment will be explained. The regression prediction used here is regression analysis in statistics, and various methods such as linear regression, nonlinear regression, and logistic regression can be applied, but the simplest case is calculated using a linear simple regression model. explain.

- In the case of traffic prediction, the traffic amount measured for each communication facility is defined as Y_k(X_k), where the measurement time point X_k is defined as time series data {(X_k, Y_k), (k=1,…,p) }. In regression analysis, we find constants a and b that define the straight line Y=aX+b that minimizes the error between time series data and the time series data. The error is generally defined as the sum of squares (least squares method).

Let the value obtained by extrapolating this straight line to future time point m be the predicted traffic value of that communication equipment Y(m)=am+b. The data length (p) to be used, etc. are included in the regression prediction calculation parameters.

- In the case of share prediction By using the share {S_k} instead of the above traffic amount {Y_k}, it is possible to perform share prediction by linear simple regression.

- For predicting the number of contracts By using the number of contracts {Z_k} instead of the above traffic volume {Y_k}, it is possible to predict the number of contracts using linear simple regression. It is also possible to predict the number of contracts for each attribute.

In S210, the traffic prediction calculation unit 16C2 acquires the attribute statistical traffic information 55 ({ST}) and the regression prediction calculation parameter ({R}) from the prediction calculation control unit 16D.

Next, in S220, the traffic prediction calculation unit 16C2 performs the above-described regression prediction calculation on the traffic using the attribute statistical traffic information 55 and the regression prediction calculation parameters. The result of the regression prediction calculation is called attribute statistical traffic prediction information 71. In S230, the attribute statistical traffic prediction information 71 is stored in the information DB unit 14.

In S310, the share prediction calculation unit 16C3 acquires the attribute statistical attribute information 56 ({AT}) and the regression prediction calculation parameter ({R}) from the prediction calculation control unit 16D.

Next, in S320, the share prediction calculation unit 16C3 performs the above-described regression prediction calculation on the attribute share using the attribute share included in the attribute statistical attribute information 56 and the regression prediction calculation parameter. The attribute share is the share (occupation rate) of users for each attribute. The result of the regression prediction calculation is called attribute statistical share prediction information 72. In S330, the attribute statistical share prediction information 72 is stored in the information DB unit 14.

Next, a method for creating the user average traffic prediction information 73 and the contract number prediction information 74 will be explained with reference to FIG. In S410, the share weighting unit 16C4 acquires the attribute statistical traffic prediction information 71 ({PT}) from the traffic prediction calculation unit 16C2, and acquires the attribute statisticalization share prediction information 72 ({PS}) from the share prediction calculation unit. Obtained from 16C3.

In S420, the share weighting unit 16C4 uses the attribute statistical traffic prediction information 71 ({PT}) and the attribute statistical share prediction information 72 ({PS}) to convert the traffic prediction by attribute into macro traffic prediction. In order to convert, share weighting processing is performed. The details of the share weighting process are as follows.

For attribute value _i (i=1,…,n) of a certain attribute item, when the predicted share value at any future point m is S_i(m) and the predicted traffic value of the attribute is TD_i(m). , the average predicted traffic value TD(m) for all target users is:

It can be calculated by The result TD(m) of the share weighting process is called user average traffic prediction information 73 ({PM}). In S430, the share weighting unit 16C4 stores the user average traffic prediction information 73 in the information DB unit 14.

In S510, the contract number prediction unit 16C5 acquires the attribute statistical attribute information 56 ({AT}) and the regression prediction calculation parameter ({R}) from the prediction calculation control unit 16C1.

Next, in S520, the number of contracts prediction unit 16C5 performs the above-described regression prediction calculation on the number of contracts using the number of contracts included in the attribute statistical attribute information 56 and the regression prediction calculation parameter. The result of the regression prediction calculation is called contract number prediction information 74 ({PU}). In S530, the contract number prediction information 74 is stored in the information DB section 14.

Next, the processing of the macro traffic growth rate prediction unit 16C6 will be explained with reference to FIG.

In S610, the macro traffic growth rate prediction unit 16C6 acquires the user average traffic prediction information 73 ({PM}) from the share weighting unit 16C4, and acquires the number of contracts prediction information 74 ({PU}) from the number of contracts prediction unit 16C5. do. Furthermore, the regression prediction calculation parameter ({R}) is read.

In S620, the macro traffic growth rate prediction unit 16C6 calculates the user average traffic volume prediction value Y(m) included in the user average traffic prediction information 73 and the contract number prediction value Z(m) contained in the contract number prediction information 74. ), Y(m)*Z(m)/Y(0) is calculated, and this is used as the macro traffic growth rate prediction information 75 ({PR}) at the future equipment design time point m.

In S630, the macro traffic growth rate prediction unit 16C6 stores the macro traffic growth rate prediction information 75 in the information DB unit 14.

(Another configuration and processing of the traffic prediction calculation unit)
Next, with reference to FIG. 6, another configuration and processing of the traffic prediction calculation unit 16C2 according to the present embodiment will be described in detail. In this example, the traffic prediction calculation unit 16C2 includes a heavy/light separation unit and a regression prediction/share weighting unit.

Even among users belonging to the same attribute, there are large differences in communication service usage and amount of transferred data consumed. Here, users with a small amount of transferred data are defined as light users, users with large amounts of transferred data are defined as heavy users, and users with the same attributes are further separated into light users and heavy users, and their respective traffic volumes and By predicting market share, we aim to improve prediction accuracy.

The characteristics of power-law distribution are used to separate light users and heavy users. Power-law distribution refers to a distribution such that the probability density function p(x|x>x _min )=Cx ^-α (α,C,x _min are positive constants). By setting both axes to logarithmic scale, it becomes a straight line with a slope (-α). It has a characteristic that the tail probability is heavier than the normal distribution.

This distribution appears in natural phenomena, and it is known that the distribution of the size of cities, the magnitude of earthquakes, computer files, wealth, etc., follows the power-law, and the amount of communication data also follows the power-law. There is a report in reference [1] (Paxson V. et al. (1995) Wide Area Traffic: The Failure of Poisson Modeling).

Fits the actual distribution, such as the amount of data transferred by users, using power-law, and separates the tail part of the distribution that follows power-law (heavy users) from the part that does not follow power-law (light users). An algorithm to do this is described in reference [2] (Clauset A. et al. (2009) Power-law Distribution in Empirical Data.).

At SS210 in FIG. 6, the traffic prediction calculation unit 16C2 performs predictive calculation control on the attribute statistical traffic information 55 ({ST}), the attribute statistical attribute information 56 ({AT}), and the regression prediction calculation parameter ({R}). 16C1.

Next, at SS220, the heavy/light separation unit separates heavy users and light users from the attribute statistical traffic information 55 using the algorithm described in Reference [2].

However, in the separation method, users who consume u times or more (u is a constant larger than 1) the average transfer data amount of the attribute determined from the attribute statistical traffic information 55 are determined to be heavy users, and those less than that are determined to be heavy users. A simple method may be used to determine that the user is a light user.

Further, the regression prediction/share weighting unit performs the following processing.

In SS230, the regression prediction/share weighting unit performs the above regression prediction calculation on the heavy user traffic (transfer data amount) using the attribute statistical traffic information 55 targeted at heavy users and the regression prediction calculation parameters. conduct.

At SS240, the regression prediction/share weighting unit calculates the heavy user's share in the relevant attribute, and performs the above-described regression prediction calculation on the share time series of the heavy user using the regression prediction calculation parameter.

In SS250, the regression prediction/share weighting unit performs the above-described regression prediction calculation on the light user's traffic (transfer data amount) using the attribute statistical traffic information 55 targeted at light users and the regression prediction calculation parameter. conduct.

At SS260, the regression prediction/share weighting unit calculates the light user's share in the relevant attribute, and performs the aforementioned regression prediction calculation on the share time series of the light user using the regression prediction calculation parameter.

At SS270, the regression prediction/share weighting unit performs share weighting processing for heavy users and light users. The obtained result is defined as attribute statistical traffic prediction information 71.

As a result, it is possible to perform traffic prediction for attributes, taking into consideration the amount of transferred data consumed and share trends of heavy users and light users. At SS280, attribute statistical traffic prediction information 71 ({PT}) is stored in the information DB unit 14.

(supplement)
For convenience of explanation, the communication band calculation device according to the present embodiment is explained using a functional block diagram, but the communication band calculation device according to the present embodiment can be implemented using hardware, software, or a combination thereof. May be realized. Further, each functional unit may be used in combination as necessary. Furthermore, the method according to this embodiment may be performed in a different order from the order shown in the embodiment.

(Effects of embodiment)
The technology described above predicts the amount of transferred data for each attribute at a future design target period based solely on statistical information based on the attributes, thereby predicting the amount of load traffic on the entire communication network being designed, and improving communications services. It becomes possible to accurately calculate the communication band required to achieve the quality of .

(Additional note)
Regarding the above embodiments, the following additional notes are further disclosed.
(Additional note 1)
A communication band calculation device that calculates the required band of communication equipment of a communication network,
memory and
at least one processor connected to the memory;
including;
The processor includes:
Obtain attribute statisticized traffic information and corresponding attribute statisticized attribute information statistically processed based on traffic information and contracted user information for each communication terminal, and traffic information for each communication equipment,
Calculating macro traffic growth rate prediction information by analysis using the attribute statistical traffic information and the attribute statistical attribute information,
A communication band calculation device that calculates a required band for each communication facility based on the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction coefficient for each communication facility.
(Additional note 2)
The processor creates statistical instruction information that is instruction information for statistical processing based on attributes from the attribute items, and transmits the statistical instruction information to a statistical processing device that executes the statistical processing. The communication band calculation device described.
(Additional note 3)
The processor includes:
Performing a regression prediction calculation on the attribute statistical traffic information to calculate attribute statistical traffic prediction information,
Performing a regression prediction calculation on the attribute statistical attribute information to calculate attribute statistical share prediction information,
performing share weighted averaging on the attribute statistical traffic prediction information and the attribute statistical share prediction information to calculate user average traffic prediction information;
Performing a regression prediction calculation on the attribute statistical attribute information to calculate contract number prediction information,
calculating the macro traffic growth rate prediction information using the user average traffic prediction information and the contract number prediction information;
The communication band calculation device according to supplementary note 1.
(Additional note 4)
The processor includes:
Separating users into heavy users and light users from the attribute statistical traffic information,
calculating the attribute statistical traffic prediction information that takes into account usage trends of heavy users and light users by performing regression prediction calculations of traffic and shares for each of heavy users and light users, and further performing share weighting;
The communication band calculation device according to Supplementary Note 3.
(Additional note 5)
A communication band calculation method executed by a computer used as a communication band calculation device for calculating the required band of communication equipment of a communication network, the method comprising:
an information acquisition step of acquiring attribute statisticized traffic information and corresponding attribute statisticized attribute information statistically processed based on traffic information and contracted user information for each communication terminal, and traffic information for each communication facility;
a prediction calculation step of calculating macro traffic growth rate prediction information by analysis using the attribute statistical traffic information and the attribute statistical attribute information;
a required bandwidth calculation step of calculating a required bandwidth for each communication facility based on the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction coefficient for each communication facility;
A communication band calculation method comprising:
(Additional note 6)
The prediction calculation step is
a traffic prediction calculation step of performing a regression prediction calculation on the attribute statistics traffic information to calculate attribute statistics traffic prediction information;
a share prediction calculation step of performing a regression prediction calculation on the attribute statistical attribute information to calculate attribute statistical share prediction information;
a share weighting step of performing share weighted averaging on the attribute statistical traffic prediction information and the attribute statistical share prediction information to calculate user average traffic prediction information;
a contract number prediction step of performing a regression prediction calculation on the attribute statistical attribute information to calculate contract number prediction information;
a macro growth rate prediction step of calculating the macro traffic growth rate prediction information using the user average traffic prediction information and the number of contracts prediction information;
The communication band calculation method according to supplementary note 5, comprising:
(Supplementary Note 7)
The traffic prediction calculation step includes:
a heavy/light separation step of separating users into heavy users and light users from the attribute statistical traffic information;
Regression prediction that calculates the attribute statistical traffic prediction information that takes into account the usage trends of heavy users and light users by performing regression prediction calculations of traffic and shares for each of heavy users and light users, and further performing share weighting.・Share weighting step,
The communication band calculation method according to supplementary note 6, comprising:
(Supplementary Note 8)
A non-temporary storage medium storing a program that causes a computer to execute each process in the communication band calculation device according to any one of Supplementary Notes 1 to 4.

As described above, by predicting the amount of transferred data for each attribute in the future design target period based only on the information made statistical by attributes, the amount of load traffic on the entire communication network to be designed can be predicted and the quality of communication services can be achieved. Although the method for accurately calculating the communication band required for the purpose of It is possible.

10 Communication band calculation device 11 Communication I/F unit 12 Operation input unit 13 Screen display unit 14 Information DB unit 15 Storage unit 16 Arithmetic processing unit 16A Information acquisition unit 16B Prediction design unit 16C Prediction calculation unit 16C1 Prediction calculation control unit 16C2 Traffic prediction Computing unit 16C3 Share prediction computing unit 16C4 Share weighting unit 16C5 Contract number prediction unit 16D Required bandwidth calculation unit 20

Communication networks

21, 22

Nodes

23, 24 Access nodes 30, 31, 32, 33, 34, 35, 36 Bandwidth equipment 41, 42, 43, 44

Communication terminals

45, 46, 47, 48 PC terminal 51 Network equipment configuration information 52 Equipment unit traffic information 53 Terminal unit traffic information 54 User unit information 55 Attribute statistics traffic information 56 Attribute statistics attribute information 57 Statistics Instruction information 61 Operation system 62 User/contract management system 63 Statistical processing device 71 Attribute statistics traffic prediction information 72 Attribute statistics share prediction information 73 User average traffic prediction information 74 Number of contracts prediction information 75 Macro traffic growth rate prediction information 76 Required bandwidth Information 77 Correction coefficient information

Claims

A communication band calculation device that calculates the required band of communication equipment of a communication network,
an information acquisition unit that acquires attribute statistical traffic information and corresponding attribute statistical attribute information statistically processed based on traffic information for each communication terminal and contracted user information, and traffic information for each communication facility;
a prediction calculation unit that calculates macro traffic growth rate prediction information by analysis using the attribute statistical traffic information and the attribute statistical attribute information;
a required bandwidth calculation unit that calculates a required bandwidth for each communication facility based on the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction coefficient for each communication facility;
A communication band calculation device comprising:
Claim further comprising: a prediction design unit that creates statistical instruction information, which is instruction information for statistical processing based on attributes, from attribute items, and transmits the statistical instruction information to a statistical processing device that executes the statistical processing. 1. The communication band calculation device according to 1.
The prediction calculation unit is
a traffic prediction calculation unit that performs regression prediction calculation on the attribute statistics traffic information to calculate attribute statistics traffic prediction information;
a share prediction calculation unit that performs a regression prediction calculation on the attribute statistical attribute information to calculate attribute statistical share prediction information;
a share weighting unit that performs share weighted averaging on the attribute statistical traffic prediction information and the attribute statistical share prediction information to calculate user average traffic prediction information;
a contract number prediction unit that performs a regression prediction calculation on the attribute statistical attribute information to calculate contract number prediction information;
a macro traffic growth rate prediction unit that calculates the macro traffic growth rate prediction information using the user average traffic prediction information and the contract number prediction information;
The communication band calculation device according to claim 1, comprising:
The traffic prediction calculation unit includes:
a heavy/light separation unit that separates users into heavy users and light users based on the attribute statistics traffic information;
Regression prediction that calculates the attribute statistical traffic prediction information that takes into account the usage trends of heavy users and light users by performing regression prediction calculations of traffic and shares for each of heavy users and light users, and further performing share weighting.・A share weighting section;
The communication band calculation device according to claim 3, comprising:
A communication band calculation method executed by a computer used as a communication band calculation device for calculating the required band of communication equipment of a communication network, the method comprising:
an information acquisition step of acquiring attribute statisticized traffic information and corresponding attribute statisticized attribute information statistically processed based on traffic information and contracted user information for each communication terminal, and traffic information for each communication facility;
a prediction calculation step of calculating macro traffic growth rate prediction information by analysis using the attribute statistical traffic information and the attribute statistical attribute information;
a required bandwidth calculation step of calculating a required bandwidth for each communication facility based on the macro traffic growth rate prediction information, the traffic information for each communication facility, and a correction coefficient for each communication facility;
A communication band calculation method comprising:
A program that causes a computer to function as each part of the communication band calculation device according to any one of claims 1 to 4.