Data recording device, data recording method, and program

Abstract

A data recording device (2) according to the present disclosure comprises: a data acquisition unit (21) for acquiring traffic data indicating the amount of traffic that has occurred for every prescribed time; a function processing unit (22) for estimating a parameter of a distribution function indicating a traffic distribution which is the distribution of traffic data in a prescribed period including a plurality of continuous prescribed times; a recording unit (23) for recording the parameter of the distribution function; and a data restoration unit (24) for, on the basis of the parameter, generating, as restored traffic data, traffic data that follows the distribution function having the parameter.

Description

[Technical Field]

[0001] The present disclosure relates to a data recording device, a data recording method, and a program. [Background technology]

[0002] A communication network is designed based on information about the bandwidth (required bandwidth) required by communication devices to transmit and receive data. In a communication network such as a public communication network, in which many users' devices are connected via lines subscribed to by the users, the communication bandwidth of each line has an upper limit (contracted bandwidth) according to the contract, but the number of lines subscribed to by each user and the contracted bandwidth of each line change over time as contracts are added, changed, or deleted. Therefore, when designing such a communication network, it is important to calculate the required bandwidth taking into account the contracted bandwidth information of each line (for example, Non-Patent Document 1).

[0003] Specifically, communication network designers determine whether a newly added line can be accommodated by determining whether the required bandwidth exceeds the bandwidth of the existing communication network (accommodation determination). If the communication network designer determines in the accommodation determination that the required bandwidth exceeds the bandwidth of the existing communication network, it becomes necessary to expand the existing facilities that make up the existing communication network. Therefore, in order to improve the efficiency of capital investment, it is important to accurately predict the traffic volume, which is the bandwidth in the communication network.

[0004] In the prior art, a data acquisition unit 91 of a bandwidth prediction device 9 as shown in Fig. 22 acquires traffic data for each interface (for each of interfaces IF1, IF2, and IF3 in the example shown in Fig. 23) from a traffic collection device 1. Then, a traffic data recording unit 92 records the traffic data, and a bandwidth prediction device 3 predicts a bandwidth based on the traffic data recorded in the traffic data recording unit 92. As shown in Fig. 23, the traffic data is data indicating, for each time, the amount of traffic occurring within a predetermined time period (five minutes in the example shown in Fig. 23) from that time. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] E. Takeshita, et al. “Traffic statistical upper limit prediction from flow features in network provisioning,” 2021 IEEE Global Communications Conference (GLOBECOM), 2021. Summary of the Invention [Problem to be solved by the invention]

[0006] However, in order to appropriately design the bandwidth of a communication network, it is necessary to use traffic data collected over a relatively long period of time. For example, traffic volume may vary depending on the season. In such cases, it is necessary to use traffic data collected over a one-year period. As shown in the example of FIG. 23, if the traffic data indicates the traffic volume generated for each time within a predetermined time (five minutes in the example of FIG. 23) from that time, the number of traffic data items per day for each line subscribed to by a user is 288. In this case, the number of traffic data items per year for each line is approximately 100,000. Furthermore, the amount of traffic data for each of the one or more lines subscribed to by many users is enormous. This poses a problem of an enormous amount of data to be recorded.

[0007] In view of the above circumstances, an object of the present disclosure is to provide a data recording device, a data recording method, and a program that can reduce the amount of data to be recorded. [Means for solving the problem]

[0008] In order to solve the above problem, the data recording device of the present disclosure includes a data acquisition unit that acquires traffic data indicating the amount of traffic generated at each predetermined time period, a function processing unit that estimates parameters of a distribution function that indicates a traffic distribution, which is the distribution of the traffic data over a predetermined period that includes multiple consecutive predetermined times, a recording unit that records the parameters, and a data restoration unit that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as restored traffic data.

[0009] In addition, in order to solve the above problem, the data recording method according to the present disclosure includes the steps of acquiring traffic data indicating the amount of traffic generated at each predetermined time interval, estimating parameters of a distribution function indicating the distribution of the traffic data over a predetermined period of time including a plurality of consecutive predetermined times, recording the parameters, and generating, based on the parameters, traffic data according to the distribution function having the parameters as restored traffic data.

[0010] In order to solve the above problem, a program according to the present disclosure causes a computer to operate as the above-mentioned data recording device. [Effects of the Invention]

[0011] According to the data recording device, data recording method, and program of the present disclosure, the amount of data to be recorded can be reduced. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a schematic diagram illustrating an example of a traffic prediction system according to a first embodiment. [Figure 2A] FIG. 10 is a diagram illustrating an example of traffic data. [Figure 2B] 2B is a histogram showing the distribution of the traffic data shown in FIG. 2A. [Figure 3] 2 is a diagram showing an example of parameters estimated by a data function processing unit shown in FIG. 1. FIG. [Figure 4A] 1. FIG. 4 is a diagram showing an example of the count number of restored traffic data generated by the data restoration unit shown in FIG. [Figure 4B] 1. FIG. 6 is a diagram showing another example of the count number of restored traffic data generated by the data restoration unit shown in FIG. [Figure 5] 2 is a diagram showing an example of restored traffic data generated by a data restoration unit shown in FIG. 1; FIG. [Figure 6]4 is a flowchart showing an operation of the data recording device shown in FIG. 1 for recording parameters of a distribution function. [Figure 7] 4 is a flowchart showing an operation of the data recording device shown in FIG. 1 to generate restored traffic data. [Figure 8] FIG. 10 is a schematic diagram illustrating an example of a traffic prediction system according to a second embodiment. [Figure 9A] FIG. 9 is a diagram showing an example of parameters of a distribution function in each predetermined period estimated by the function processing unit shown in FIG. 8. [Figure 9B] FIG. 9 is a diagram showing an example of parameters of a distribution function synthesized by a function synthesis unit shown in FIG. 8. [Figure 10] 9 is a flowchart showing an operation of the data recording device shown in FIG. 8 to record parameters of a distribution function. [Figure 11] FIG. 10 is a schematic diagram illustrating an example of a traffic prediction system according to a third embodiment. [Figure 12A] 12 is a diagram showing an example of parameters of a distribution function estimated by a function processing unit shown in FIG. 11 and outlier traffic data determined by an outlier extraction unit. FIG. [Figure 12B] 12 is a diagram showing an example of restored traffic data generated by a data restoration unit shown in FIG. 11. FIG. [Figure 13] 12 is a flowchart showing an operation of the data recording device shown in FIG. 11 for recording parameters of a distribution function. [Figure 14] FIG. 10 is a schematic diagram illustrating an example of a traffic prediction system according to a fourth embodiment. [Figure 15A] FIG. 2 is a diagram showing an example of a histogram indicating the distribution of traffic data and a first distribution function formed by one distribution function. [Figure 15B] FIG. 10 is a diagram showing an example of a histogram indicating the distribution of traffic data and a second distribution function obtained by superimposing two distribution functions. [Figure 15C] FIG. 10 is a diagram showing an example of a histogram showing the distribution of traffic data and a third distribution function obtained by superimposing three distribution functions. [Figure 16]15 is a flowchart showing an operation of the data recording device shown in FIG. 14 for recording parameters of a distribution function. [Figure 17] FIG. 10 is a schematic diagram illustrating an example of a traffic prediction system according to a fifth embodiment. [Figure 18] 18 is a flowchart showing an operation of the data recording device shown in FIG. 17 for recording parameters of a distribution function. [Figure 19] FIG. 10 is a schematic diagram illustrating an example of a traffic prediction system according to a sixth embodiment. [Figure 20] 20 is a flowchart showing an operation of the data recording device shown in FIG. 19 for recording parameters of a distribution function. [Figure 21] FIG. 2 is a diagram illustrating an example of a hardware configuration of the data recording device illustrated in FIG. [Figure 22] FIG. 1 is a schematic diagram illustrating a conventional traffic prediction system. [Figure 23] FIG. 10 is a diagram illustrating an example of traffic data for each interface. DETAILED DESCRIPTION OF THE INVENTION

[0013] <<First embodiment>> The overall configuration of the first embodiment will be described with reference to Fig. 1. Fig. 1 is a schematic diagram showing an example of a traffic prediction system 100 according to the first embodiment.

[0014] The traffic prediction system 100 includes a traffic collection device 1, a data recording device 2, and a bandwidth prediction device 3. The traffic collection device 1 and the data recording device 2 communicate with each other via a communication network. The data recording device 2 and the bandwidth prediction device 3 also communicate with each other via the communication network. The traffic prediction system 100 may further include a line information storage device (omitted in FIG. 1). The line information storage device stores line information and can transmit the line information to the data recording device 2 via the communication network. The line information storage device and the line information will be described in detail in the second embodiment.

[0015] <Traffic collection device> The traffic collection device 1 is configured by a computer including a memory, a controller, and a communication interface. The memory may be configured by a hard disk drive (HDD), a solid state drive (SSD), an electrically erasable programmable read-only memory (EEPROM), a read-only memory (ROM), a random access memory (RAM), or the like. The controller may be configured by dedicated hardware such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA), or may be configured by a processor, or may be configured by including both. The communication interface may use standards such as Ethernet (registered trademark), FDDI (Fiber Distributed Data Interface), or Wi-Fi (registered trademark).

[0016] The traffic collection device 1 collects traffic data. As shown in FIG. 2A, the traffic data is data indicating the amount of traffic generated at each predetermined time interval. In the example shown in FIG. 1, the predetermined time interval is 5 minutes. For example, it is shown that 31 MB of traffic occurred in the 5 minutes from 00:00 on January 1, 2022, and 40 MB of traffic occurred in the 5 minutes from 23:55 on January 1, 2022. The traffic collection device 1 also records the collected traffic data. Note that, in the example shown in FIG. 2A, the predetermined time interval is 5 minutes, but is not limited to this. Also, in FIG. 2A, January 1, 2022 is shown as "2022 / 1 / 1," and the same is true for other dates and other figures.

[0017] <Data recording device> As shown in Fig. 1, the data recording device 2 includes a data acquisition unit 21, a function processing unit 22, a recording unit 23, and a data restoration unit 24. The data acquisition unit 21 is configured by an input interface. The input interface may be a communication interface. The function processing unit 22 and the data restoration unit 24 are configured by a controller. The recording unit 23 is configured by a memory.

[0018] The data acquiring unit 21 acquires, at any timing, traffic data indicating the amount of traffic generated at each predetermined time interval from the traffic collection device 1 via the communication network. The data acquiring unit 21 also outputs the acquired traffic data to the function processing unit 22.

[0019] The function processing unit 22 estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of traffic data over a predetermined period of time that includes a plurality of consecutive predetermined times. The predetermined period is a period that is set in advance by the operator of the data recording device 2. In the example shown in FIG. 2A, the predetermined period is one day, and the traffic data over the predetermined period (one day) is surrounded by a dashed line. Note that, although the predetermined period is one day in the example shown in FIG. 2A, it is not limited to this.

[0020] Specifically, the function processing unit 22 generates a traffic data histogram as shown in Fig. 2B for each predetermined period. The traffic data histogram indicates the number of times (count number) that a traffic volume included in a traffic volume range (traffic width) occurred. The traffic width can be calculated as log2 (number of traffic data) + 1 according to Sturges's formula, for example.

[0021] In the example shown in Fig. 2B, in the histogram of traffic data, the horizontal axis indicates the traffic width, and the vertical axis indicates the count number of occurrences of each traffic volume on January 1, 2022. Although Fig. 2B shows a histogram related to the traffic volume occurring on January 1, 2022, the function processing unit 22 may similarly generate histograms for other predetermined periods.

[0022] Then, the function processing unit 22 estimates the parameters of the distribution function by fitting a distribution function (see the dashed line in FIG. 2B) to the distribution of traffic data (traffic distribution) shown in the histogram. For example, the function processing unit 22 may determine the parameters of the distribution function in the histogram using the least squares method. Alternatively, the function processing unit 22 may determine the parameters of the distribution function that maximizes the likelihood by using a likelihood function, which is a function obtained by substituting each piece of traffic data into the distribution function and multiplying the results. The distribution function may be determined by an administrator of the data recording device 2 or the like based on the characteristics of the traffic volume of the target communication network. The distribution function may be, for example, a function indicating a normal distribution. In this case, the parameters of the distribution function may be the mean value μ and standard deviation σ of the traffic data.

[0023] Furthermore, the function processing unit 22 associates the determined parameters of the distribution function with a predetermined period and records them in the recording unit 23. In the example shown in Fig. 3, the function processing unit 22 associates the parameters, an average value μ of "37M" bytes and a standard deviation of "5M" bytes, with the predetermined period, "January 1, 2022," and records them in the recording unit 23.

[0024] The recording unit 23 records the parameters estimated by the function processing unit 22. The recording unit 23 may record the parameters of the distribution function in association with a predetermined period, as shown in FIG.

[0025] Based on the parameters recorded in the recording unit 23, the data restoration unit 24 generates traffic data according to a distribution function having the parameters as restored traffic data. Specifically, the data restoration unit 24 acquires the parameters of the distribution function recorded in the recording unit 23. Then, the data restoration unit 24 calculates the count number for each traffic width based on the distribution function having the parameters. The number of restored traffic data generated by the data restoration unit 24 may be any number, but may be the same as the number of traffic data used to estimate the parameters, for example.

[0026] 4A using random numbers to form a histogram according to a distribution function, and generate restored traffic data of the counts corresponding to each traffic width. For example, the data restoration unit 24 may use the norm.inv function in the spreadsheet software "Excel (registered trademark)" or the np.random.normal function in the programming language "Python (registered trademark)."

[0027] As another example, the data restoration unit 24 may calculate, for each traffic width, a count number corresponding to the value of the distribution function corresponding to the traffic width, as indicated by the circle in FIG. 4B, and generate restored traffic data of the count number corresponding to each traffic width.

[0028] The restored traffic data generated by the data restoration unit 24 in this manner is data indicating the amount of traffic that occurred at predetermined times (five minutes in this example) during a predetermined period (one day in this example), as shown in Fig. 5. In the example shown in Fig. 5, 288 pieces of restored traffic data are shown, each indicating that traffic amounts of 31 MB, . . . , and 45 MB occurred during each five-minute period on January 1, 2022.

[0029] Furthermore, the data restoration unit 24 outputs the restored traffic data to the bandwidth prediction device 3.

[0030] <Bandwidth prediction device> The bandwidth prediction device 3 is configured by a computer equipped with a memory, a controller, and a communication interface. The bandwidth prediction device 3 acquires the restored traffic data output from the data recording device 2. The bandwidth prediction device 3 also predicts the bandwidth based on the restored traffic data using any method.

[0031] <Operation of the data recording device> Next, the operation of the data recording device 2 according to the first embodiment will be described with reference to Fig. 6 and Fig. 7. Fig. 6 and Fig. 7 are sequence diagrams showing an example of the operation of the data recording device 2 according to this embodiment. The operation of the data recording device 2 described with reference to Fig. 6 corresponds to a data recording method executed by the data recording device 2 according to this embodiment.

[0032] First, with reference to FIG. 6, the operation of the data recording device 2 for recording the parameters of the distribution function will be described.

[0033] In step S11, the data acquisition unit 21 acquires traffic data from the traffic collection device 1 via the communication network.

[0034] In step S12, the function processing unit 22 estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of traffic data for a predetermined period. Specifically, the function processing unit 22 generates a histogram of the traffic data for each predetermined period. Then, the function processing unit 22 estimates parameters of the distribution function by fitting the distribution function to the traffic distribution indicated by the histogram.

[0035] In step S13, the recording unit 23 records the parameters of the distribution function estimated by the function processing unit 22.

[0036] Next, with reference to FIG. 7, an operation of the data recording device 2 for generating restored traffic data based on the parameters of the distribution function will be described.

[0037] In step S14, the data restoration unit 24 acquires the parameters of the distribution function recorded in the recording unit 23.

[0038] In step S15, the data restoration unit 24 calculates the count number for each traffic width based on the distribution function determined by the parameters.

[0039] In step S16, the data restoration unit 24 generates restored traffic data indicating the traffic volume included in each traffic width, for the number of counts in each traffic width.

[0040] As described above, according to the first embodiment, the data recording device 2 includes a data acquiring unit 21 that acquires traffic data indicating the amount of traffic generated at each predetermined time interval, a function processing unit 22 that estimates parameters of a distribution function that indicates the distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times, a recording unit 23 that records the parameters, and a data restoring unit 24 that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as restored traffic data.

[0041] As a result, the data recording device 2 records only the parameters of the distribution function, rather than the traffic data itself, and therefore the amount of data to be recorded can be reduced. For example, if the distribution function is a function indicating a normal distribution and traffic data indicating traffic volume every five minutes is acquired for one day, the number of traffic data will be 288. In this case, with the above-mentioned configuration, the data recording device 2 only needs to record two data items, the mean value μ and the standard deviation σ, which are parameters of the distribution function. Therefore, the data recording device 2 can reduce the amount of data to be recorded.

[0042] Furthermore, according to the first embodiment, the data recording device 2 can generate any number of restored traffic data. This allows the data recording device 2 to generate the number of restored traffic data required for the desired accuracy in bandwidth prediction by the bandwidth prediction device 3. Therefore, the data recording device 2 can realize bandwidth prediction by the bandwidth prediction device 3 with the desired accuracy.

[0043] <<Second embodiment>> The overall configuration of the second embodiment will be described with reference to Fig. 8. Fig. 8 is a schematic diagram showing an example of a traffic prediction system 100-1 according to the second embodiment.

[0044] The traffic prediction system 100-1 includes a traffic collection device 1, a data recording device 2-1, a bandwidth prediction device 3, and a line information storage device 4. The traffic collection device 1 and the line information storage device 4 communicate with the data recording device 2-1 via a communication network. The data recording device 2-1 and the bandwidth prediction device 3 also communicate with each other via a communication network. In the second embodiment, the same functional units as in the first embodiment are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0045] The line information storage device 4 is configured by a computer having a memory, a controller, and a communication interface. The line information storage device 4 stores line information. The line information is any information related to the line, and can be, for example, the contracted bandwidth. The line information may be, for example, a user name, a service used, configuration information related to communication (router, network, etc.), or information related to a protocol. In the second embodiment, an example is described in which each functional unit executes processing using the contracted bandwidth, which is an example of line information, but each functional unit may execute processing using any information related to the line, not limited to the contracted bandwidth. The same applies to the sixth embodiment, which will be described later.

[0046] The data recording device 2-1 includes a data acquisition unit 21-1, a function processing unit 22-1, a recording unit 23-1, a data restoration unit 24, a determination unit 25, and a function synthesis unit 26. The data acquisition unit 21-1 is configured by an input interface. The function processing unit 22-1, the determination unit 25, and the function synthesis unit 26 are configured by a controller. The recording unit 23-1 is configured by a memory.

[0047] Similar to the data acquiring unit 21 of the first embodiment, the data acquiring unit 21-1 acquires traffic data indicating the amount of traffic generated at predetermined time intervals from the traffic collection device 1 via the communication network at any timing. The data acquiring unit 21-1 also acquires line information corresponding to the traffic data from the line information storage device 4 via the communication network at any timing. Furthermore, the data acquiring unit 21-1 associates the traffic data with line information corresponding to the line of the traffic data and outputs the associated data to the function processing unit 22-1.

[0048] The function processing unit 22-1 estimates parameters of a distribution function that indicates a traffic distribution for each predetermined period. Specifically, the function processing unit 22-1 estimates a first parameter calculated based on traffic data for a first predetermined period, and, with the first parameter recorded in the recording unit 23-1, estimates a second parameter based on traffic data for a second predetermined period that follows the first predetermined period. The predetermined period in this embodiment is a period during which the contracted bandwidth does not change, and can be, for example, one day. The specific processing by the function processing unit 22-1 is the same as the specific processing by the function processing unit 22 in the first embodiment.

[0049] The recording unit 23-1 records line information (e.g., contracted bandwidth) and distribution function parameters for a predetermined period in association with the predetermined period (e.g., the date of the predetermined period). The recording unit 23-1 also records line information and composite parameters for the predetermined period in association with the predetermined period. For example, in the example shown in FIG. 9B, the recording unit 23-1 records composite parameters, such as an average value μ of 37 Mbytes and a standard deviation of 5 Mbytes, and line information, such as contracted bandwidths of 100 Mbytes and 200 Mbytes, in association with the predetermined period "2020 / 1 / 1." The recording unit 23-1 also records distribution function parameters, such as an average value μ of 50 Mbytes and a standard deviation of 8 Mbytes, and line information, such as contracted bandwidths of 100 Mbytes and 300 Mbytes, in association with the predetermined period "2020 / 1 / 3." The composite parameters recorded here will be described in detail later.

[0050] The determination unit 25 determines whether the line information for the second predetermined period is the same as the line information for the first predetermined period, and if it determines that the contracted bandwidth of the line for the second predetermined period is the same as the line information for the first predetermined period, it determines whether the difference between the first parameter and the second parameter is equal to or less than a first threshold. In an example where the line information is the contracted bandwidth, if multiple lines are contracted for the predetermined period, the determination unit 25 determines whether the line information for each of the multiple lines is the same.

[0051] In the example shown in FIG. 9A, the first predetermined period is "January 1, 2022," and the second predetermined period is "January 2, 2022." For example, the above-mentioned function processing unit 22-1 estimates a parameter (first parameter) of the distribution function based on traffic data for the first predetermined period, "January 1, 2022," and the recording unit 23-1 records the first parameter. In this state, the function processing unit 22-1 estimates a parameter (second parameter) of the distribution function based on traffic data for the second predetermined period, "January 2, 2022." Then, the determination unit 25 determines whether the contracted bandwidths (100 MB and 200 MB), which are the line information for the second predetermined period, are the same as the contracted bandwidths (100 MB and 200 MB), which are the line information for the first predetermined period. In this example, the function processing unit 22-1 determines that the contracted bandwidths are the same.

[0052] If the determination unit 25 determines that the line information for the first predetermined period is the same as the line information for the second predetermined period, the determination unit 25 determines whether the difference between the first parameter and the second parameter is equal to or less than a first threshold. The first threshold is a value set by the operator of the data recording device 2-1. The first threshold is a value at which the difference between the prediction results by the bandwidth prediction device 3 when the traffic distributions for the two predetermined periods are shown together and when the traffic distributions for the two predetermined periods are shown separately is considered to be an error.

[0053] In the example shown in FIG. 9A, the average value μ, which is the first parameter, is "36M" bytes, and the average value μ, which is the second parameter, is "38M" bytes. Therefore, the difference is "2M" bytes. As a result, when the first threshold is "5M" bytes, the determination unit 25 determines that the difference is equal to or less than the first threshold. Furthermore, the standard deviation σ, which is the first parameter, is "4M" bytes, and the standard deviation σ, which is the second parameter, is "6M" bytes. Therefore, the difference is "2M" bytes. As a result, when the first threshold is "5M" bytes, the determination unit 25 determines that the difference is equal to or less than the first threshold.

[0054] If it is determined that the difference is equal to or smaller than the first threshold, the function synthesis unit 26 calculates a synthesis parameter based on the first parameter and the second parameter. Specifically, the function synthesis unit 26 calculates the average value of the first parameter and the second parameter as the synthesis parameter.

[0055] In the example shown in Fig. 9A, the average value μ as a first parameter is "36M" bytes, and the average value μ as a second parameter is "38M" bytes, so the function synthesis unit 26 calculates "37M" bytes, which is the average of these, as the synthesis parameter, as shown in Fig. 9B. Furthermore, the standard deviation σ as a first parameter is "4M" bytes, and the standard deviation σ as a second parameter is "6M" bytes, so the function synthesis unit 26 calculates "5M" bytes, which is the average of these, as the synthesis parameter, as shown in Fig. 9B.

[0056] When the synthesis parameter is calculated, the function synthesis unit 26 causes the recording unit 23-1 to record the synthesis parameter and line information for each period obtained by combining the first predetermined period and the second predetermined period.

[0057] If the determining unit 25 determines that the line information differs, the function combining unit 26 does not combine the traffic distribution for the first predetermined period with the traffic distribution for the second predetermined period. If the determining unit 25 determines that the difference is greater than the first threshold, the function combining unit 26 does not combine the traffic distribution for the first predetermined period with the traffic distribution for the second predetermined period. In this case, the function combining unit 26 causes the recording unit 23-1 to record the parameters and line information for each predetermined period.

[0058] <Operation of the data recording device> Next, the operation of the data recording device 2-1 according to the second embodiment will be described with reference to Fig. 10. Fig. 10 is a sequence diagram showing an example of the operation of the data recording device 2-1 according to the second embodiment. The operation of the data recording device 2-1 described with reference to Fig. 10 corresponds to a data recording method executed by the data recording device 2-1 according to the second embodiment. In the example shown in Fig. 10, at the start of the operation, the function processing unit 22-1 calculates a first parameter based on traffic data for a first predetermined period, and the recording unit 23-1 records the first parameter. In addition, line information for the first predetermined period is acquired.

[0059] In step S21, the data acquisition unit 21-1 acquires traffic data from the traffic collection device 1 via the communication network.

[0060] In step S22, the data acquisition unit 21-1 acquires line information for each line subscribed to by the user from the line information storage device 4 via the communication network.

[0061] In step S23, the function processing unit 22-1 estimates a second parameter which is a parameter indicating the distribution function for a second predetermined period after the first predetermined period.

[0062] In step S24, the determination unit 25 acquires the first parameter and the line information for the first predetermined period from the recording unit 23.

[0063] In step S25, the determination unit 25 determines whether the line information in the second predetermined period is the same as the line information in the first predetermined period.

[0064] If, in step S25, the judgment unit 25 judges that the line information for the first predetermined period is not identical to the line information for the second predetermined period, in step S26, the recording unit 23-1 records the first parameter and the second parameter, respectively.

[0065] If it is determined in step S25 that the line information for the first predetermined period is the same as the line information for the second predetermined period, then in step S27, the determination unit 25 determines whether the difference between the first parameter and the second parameter is less than or equal to a first threshold value.

[0066] If it is determined in step S27 that the difference is greater than the first threshold value, then in step S26, the recording unit 23-1 records the first parameter and the second parameter.

[0067] If it is determined in step S27 that the difference is equal to or smaller than the first threshold, in step S28, the function synthesis unit 26 calculates a synthesis parameter based on the first parameter and the second parameter.

[0068] In step S29, the recording unit 23-1 records the composite parameter. Specifically, the recording unit 23-1 updates the first parameter recorded in the recording unit 23-1 to the composite parameter.

[0069] The operation of the data recording device 2-1 in the second embodiment to generate restored traffic data is similar to the operation of the data recording device 2 in the first embodiment to generate restored traffic data.

[0070] As described above, according to the second embodiment, the data recording device 2-1 further includes a determination unit 25 and a function synthesis unit 26, the data acquisition unit 21-1 acquires line information corresponding to traffic data, the function processing unit 22-1 estimates a first parameter which is a parameter indicating a distribution function for a first predetermined period, and, with the first parameter recorded in the recording unit 23-1, estimates a second parameter which is a parameter indicating a distribution function for a second predetermined period after the first predetermined period, the determination unit 25 determines whether the line information for the second predetermined period is identical to the line information for the first predetermined period, and if it is determined that the line information for the second predetermined period is identical to the line information for the first predetermined period, determines whether the difference between the first parameter and the second parameter is equal to or less than a first threshold, and if it is determined that the difference is equal to or less than the first threshold, the function synthesis unit 26 calculates a synthesis parameter based on the first parameter and the second parameter, and the recording unit 23-1 records the synthesis parameter.

[0071] In this way, the data recording device 2-1 records composite parameters, which are parameters of a function that indicates the distribution of traffic data over multiple predetermined periods, thereby further reducing the amount of data to be recorded compared to a configuration that records parameters for each predetermined period. For example, if the distribution function is a function that indicates a normal distribution and the predetermined period is one day, the data recording device 2 of the first embodiment records two parameters (average value μ and standard deviation σ) for each day for 365 days, thereby recording a total of 730 parameters. In contrast, if the line information remains the same for one year, the data recording device 2-1 of the second embodiment only needs to record two parameters: a composite parameter obtained by combining the average values ​​μ for 365 days and a composite parameter obtained by combining the standard deviations σ for 365 days. In this way, the data recording device 2-1 can further reduce the amount of data to be recorded.

[0072] <<Third embodiment>> The overall configuration of the third embodiment will be described with reference to Fig. 11. Fig. 11 is a schematic diagram showing an example of a traffic prediction system 100-2 according to the third embodiment.

[0073] The traffic prediction system 100-2 includes a traffic collection device 1, a data recording device 2-2, and a bandwidth prediction device 3. The traffic collection device 1 and the data recording device 2-2 communicate with each other via a communication network. The data recording device 2-2 and the bandwidth prediction device 3 also communicate with each other via a communication network. The traffic prediction system 100-2 may further include a line information storage device (omitted in FIG. 11). The line information storage device stores line information and can transmit the line information to the data recording device 2-2 via the communication network. The line information storage device and the line information have been described in detail in the second embodiment. In the third embodiment, functional units that are the same as those in the first embodiment are assigned the same reference numerals, and descriptions thereof will be omitted.

[0074] The data recording device 2-2 includes a data acquisition unit 21, a function processing unit 22, a recording unit 23-2, a data restoration unit 24-2, and an outlier extraction unit 27. The data restoration unit 24-2 and the outlier extraction unit 27 are configured by a controller. The recording unit 23-2 is configured by a memory.

[0075] The outlier extraction unit 27 extracts deviant traffic data from the traffic data acquired by the data acquisition unit 21. The deviant traffic data is traffic data that deviates from the traffic distribution indicated by the distribution function having the parameters estimated by the function processing unit 22. The outlier extraction unit 27 may extract traffic data that is not within a set range centered on the average value μ as the deviant traffic data. The set range can be arbitrarily set by an administrator of the data recording device 2-2, and can be, for example, the range between the value obtained by subtracting three times the standard deviation σ from the average value μ and the value obtained by adding three times the standard deviation σ to the average value μ.

[0076] Furthermore, the outlier extractor 27 causes the recorder 23-2 to record the outlier traffic data.

[0077] As shown in Fig. 12A, the recording unit 23-2 records the parameters of the distribution function determined by the function processing unit 22, similar to the recording unit 23 of the first embodiment. The recording unit 23-2 further records deviating traffic data. In the example shown in Fig. 12A, the recording unit 23-2 stores deviating traffic volume 1 and deviating traffic volume 2 as the deviating traffic data, in association with a time indicating a predetermined period.

[0078] 12B, the data restoration unit 24-2 generates traffic data according to a traffic distribution represented by a distribution function having estimated parameters and deviant traffic data as restored traffic data. The specific process performed by the data restoration unit 24-2 to generate traffic data according to a traffic distribution represented by a distribution function having estimated parameters is similar to the process performed by the data restoration unit 24 of the first embodiment to generate restored traffic data.

[0079] At this time, the data restoration unit 24-2 may generate traffic data according to a traffic distribution represented by a distribution function having estimated parameters, the number of which is obtained by subtracting the number of deviant traffic data from the total number of restored traffic data. In the example described in the first embodiment, the data restoration unit 24-2 generates 288 pieces of restored traffic data, but in the second embodiment, for example, if the number of deviant traffic data is 2, the data restoration unit 24-2 may generate 286 (288-2) pieces of traffic data according to a traffic distribution represented by a distribution function having estimated parameters.

[0080] <Operation of the data recording device> Next, the operation of the data recording device 2-2 according to the third embodiment will be described with reference to Fig. 13. Fig. 13 is a sequence diagram showing an example of the operation of the data recording device 2-2 according to the third embodiment. The operation of the data recording device 2-2 described with reference to Fig. 13 corresponds to the data recording method executed by the data recording device 2-2 according to the third embodiment.

[0081] In step S31, the data acquisition unit 21 acquires traffic data from the traffic collection device 1 via the communication network.

[0082] In step S32, the function processing unit 22 estimates parameters of a distribution function that indicates a traffic distribution, which is the distribution of traffic data for a predetermined period of time.

[0083] In step S33, the recording unit 23-2 records the parameters of the distribution function estimated by the function processing unit 22.

[0084] In step S34, the outlier extractor 27 extracts outlier traffic data from the traffic data acquired by the data acquirer 21.

[0085] In step S35, the recording unit 23-2 further records the off-traffic data.

[0086] In addition, the operation of the data recording device 2-2 in the third embodiment to generate restored traffic data is the same as the operation of the data recording device 2 in the first embodiment to generate restored traffic data, plus the operation of adding off-track traffic data as restored traffic data.

[0087] As described above, according to the third embodiment, the data recording device 2-2 further includes an outlier extraction unit 27 that extracts deviant traffic data, which is traffic data that deviates from the traffic distribution indicated by the distribution function having the parameters estimated by the function processing unit 22, from the traffic data acquired by the data acquisition unit 21. The recording unit 23-2 further records the deviant traffic data, and the data restoration unit 24-2 generates restored traffic data from the traffic data that follows the traffic distribution indicated by the distribution function having the above parameters and the deviant traffic data.

[0088] As a result, the data recording device 2-2 can generate restored traffic data that reproduces the traffic data collected by the traffic collecting device 1 with higher accuracy. If the traffic data collected by the traffic collecting device 1 includes outlier traffic data, the data restoration unit 24 of the data recording device 2 of the first embodiment cannot generate restored traffic data equivalent to the outlier traffic data. This can cause a discrepancy between the traffic data collected by the traffic collecting device 1 and the restored traffic data. In contrast, the data recording device 2-2 of the third embodiment, with the above-described configuration, can generate restored traffic data equivalent to the outlier traffic data. Therefore, even if an outlier exists in the traffic data, it is possible to suppress a discrepancy between the traffic data collected by the traffic collecting device 1 and the restored traffic data.

[0089] <<Fourth embodiment>> The overall configuration of the fourth embodiment will be described with reference to Fig. 14. Fig. 14 is a schematic diagram showing an example of a traffic prediction system 100-3 according to the fourth embodiment.

[0090] The traffic prediction system 100-3 includes a traffic collection device 1, a data recording device 2-3, and a bandwidth prediction device 3. The traffic collection device 1 and the data recording device 2-3 communicate with each other via a communication network. The data recording device 2-3 and the bandwidth prediction device 3 also communicate with each other via the communication network. The traffic prediction system 100-3 may further include a line information storage device (omitted in FIG. 14). The line information storage device stores line information and can transmit the line information to the data recording device 2-3 via the communication network. The line information storage device and the line information have been described in detail in the second embodiment. In the fourth embodiment, the same functional units as in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[0091] <Data recording device> The data recording device 2-3 includes a data acquisition unit 21, a function processing unit 22-3, a recording unit 23, a data restoration unit 24, and an overlay determination unit 28. The function processing unit 22-3 and the overlay determination unit 28 are configured by a controller.

[0092] The function processing unit 22-3 estimates each parameter of a superposition distribution function including n distribution functions (n ​​is an integer equal to or greater than 1) for each predetermined period. When n is 1, the superposition distribution function is one distribution function, and when n is an integer equal to or greater than 2, the superposition distribution function is a function obtained by superposing n distribution functions.

[0093] The overlap determination unit 28 determines whether the error (e.g., mean square error) between the traffic distribution indicated by the overlap distribution function having each estimated parameter and the traffic distribution of the traffic data acquired by the data acquisition unit 21 is less than or equal to a second threshold.

[0094] The second threshold is a value that is set in advance by an administrator or the like of the data recording device 2. The smaller the second threshold, the smaller the deviation between the traffic distribution indicated by the distribution function having the estimated parameters and the traffic distribution acquired by the data acquiring unit 21. However, the processing load and the amount of parameters to be recorded increase due to the increase in the number of distribution functions to be superimposed. In other words, the larger the second threshold, the larger the deviation between the traffic distribution indicated by the distribution function having the estimated parameters and the traffic distribution acquired by the data acquiring unit 21. However, the smaller the number of distribution functions to be superimposed, the smaller the processing load and the amount of parameters to be recorded can be suppressed. Therefore, it is preferable that the second threshold be determined depending on the deviation of the traffic distribution, the processing load, and the amount of parameters to be recorded.

[0095] Then, when it is determined that the error is equal to or less than the second threshold, the overlay determination unit 28 causes the recording unit 23 to record the parameters of the n-th distribution function. That is, when it is determined that the error is equal to or less than the second threshold, the recording unit 23 records the parameters of the overlay distribution function. When it is determined that the error is greater than the second threshold, the function processing unit 22-3 sets n=n+1 and again estimates the parameters of the overlay distribution function obtained by overlaying n distribution functions.

[0096] 15A, 15B, and 15C, the functions of the function processing unit 22-3 and the overlap determination unit 28 will be described in detail. Note that, hereinafter, the "nth (n is an integer of 1 or more) distribution function" refers to a distribution function including n distribution functions. Specifically, the "first distribution function" refers to one distribution function, and the "nth (n is an integer of 2 or more) distribution function" refers to a function obtained by overlapping n distribution functions.

[0097] The function processing unit 22-3 estimates parameters of a first distribution function (see FIG. 15A) that indicates a traffic distribution, which is the distribution of traffic data for a predetermined period of time.

[0098] The overlap determining unit 28 determines whether the error between the first distribution function having the estimated parameters and the traffic data is equal to or less than a second threshold value.

[0099] When it is determined that the error between the first distribution function having the estimated parameters and the traffic data is equal to or smaller than the second threshold, the function processing unit 22-3 causes the recording unit 23 to record the parameters of the first distribution function.

[0100] If it is determined that the error is greater than the second threshold, the overlap determination unit 28 estimates the parameters of a second distribution function (see FIG. 15B) obtained by overlapping the two distribution functions. Then, the function processing unit 22-3 determines whether the error between the second distribution function having the estimated parameters and the traffic data is equal to or less than the second threshold.

[0101] When it is determined that the error between the second distribution function having the estimated parameters and the traffic data is equal to or less than the second threshold, the function processing unit 22-3 causes the recording unit 23 to record the parameters of the second distribution function. Then, the recording unit 23 records the parameters of the second distribution function.

[0102] If it is determined that the error is greater than the second threshold, the overlap determination unit 28 estimates the parameters of a third distribution function (see FIG. 15C) obtained by overlapping the three distribution functions. Then, the function processing unit 22-3 determines whether the error between the third distribution function having the estimated parameters and the traffic data is equal to or less than the second threshold.

[0103] When it is determined that the error between the traffic data and the third distribution function having the estimated parameters is equal to or less than the third threshold, the function processing unit 22-3 causes the recording unit 23 to record the parameters of the third distribution function. Then, the recording unit 23 records the parameters of the third distribution function.

[0104] In the fourth embodiment, the nth distribution function may be a function obtained by superposing n distribution functions each showing a normal distribution, as shown in formula (1). In formula (1), x is the traffic volume, and y is the count number in the histogram of the traffic data. i , μ i , σ i are the coefficient, mean value, and standard deviation, respectively, of the ith distribution function among the n distribution functions superimposed in the nth distribution function.

[0105]

number

[0106] <Operation of the data recording device> Next, the operation of the data recording device 2-3 according to the fourth embodiment will be described with reference to Fig. 16. Fig. 16 is a sequence diagram showing an example of the operation of the data recording device 2-3 according to the fourth embodiment. The operation of the data recording device 2-3 described with reference to Fig. 16 corresponds to the data recording method executed by the data recording device 2-3 according to the fourth embodiment.

[0107] In step S41, the data acquisition unit 21 acquires traffic data from the traffic collection device 1 via the communication network.

[0108] In step S42, the function processing unit 22-3 sets n=1.

[0109] In step S43, the function processing unit 22-3 estimates the parameters of the n-th distribution function including n distribution functions for each predetermined period.

[0110] In step S44, the overlap determination unit 28 determines whether the error between the traffic distribution indicated by the nth distribution function having the estimated parameters and the traffic distribution of the traffic data acquired by the data acquisition unit 21 is less than or equal to a second threshold value.

[0111] If it is determined in step S44 that the error is greater than the second threshold value, then in step S45, the function processing unit 22-3 sets n = n + 1. Then, the data recording device 2-3 returns to step S43.

[0112] If it is determined in step S44 that the error is equal to or less than the second threshold, then in step S46, the recording unit 23 records the parameters of the n-th distribution function.

[0113] The operation of the data recording device 2-3 in the fourth embodiment to generate restored traffic data is similar to the operation of the data recording device 2 in the first embodiment to generate restored traffic data.

[0114] As described above, according to the fourth embodiment, the data recording device 2-3 further includes an overlap determination unit 28, and the function processing unit 22-3 estimates an overlap distribution function including n (n is an integer equal to or greater than 1) distribution functions for each predetermined period. The overlap determination unit 28 determines whether the error between the traffic distribution indicated by the overlap distribution function and the traffic distribution of the traffic data acquired by the data acquiring unit 21 is equal to or less than a second threshold. If the recording unit 23 determines that the error is equal to or less than the second threshold, it records the parameters of the overlap distribution function. If the function processing unit 22-3 determines that the error is greater than the second threshold, it sets n=n+1 and again estimates the parameters of the overlap distribution function including n distribution functions.

[0115] This allows the data recording device 2-3 to record parameters of the superposition distribution function that reduce the error between the traffic distribution of the traffic data acquired by the data acquiring unit 21 and the traffic data. This allows the data recording device 2-3 to suppress the deviation between the traffic data collected by the traffic collecting device 1 and the restored traffic data. In other words, the data recording device 2-3 can improve the reproducibility of the restored traffic data.

[0116] <<Fifth embodiment>> The overall configuration of the fifth embodiment will be described with reference to Fig. 17. Fig. 17 is a schematic diagram showing an example of a traffic prediction system 100-4 according to the fifth embodiment.

[0117] The traffic prediction system 100-4 includes a traffic collection device 1, a data recording device 2-4, and a bandwidth prediction device 3. The traffic collection device 1 and the data recording device 2-4 communicate with each other via a communication network. The data recording device 2-4 and the bandwidth prediction device 3 also communicate with each other via the communication network. The traffic prediction system 100-4 may further include a line information storage device (omitted in FIG. 17). The line information storage device stores line information and can transmit the line information to the data recording device 2-4 via the communication network. The line information storage device and the line information are described in detail in the second embodiment. In the fifth embodiment, functional units that are the same as those in the first embodiment are assigned the same reference numerals, and descriptions thereof will be omitted.

[0118] The data recording device 2-4 includes a data acquisition unit 21, a plurality of function processing units 22-4k (k=1 to m, k and m are integers equal to or greater than 1), a recording unit 23-4, a data restoration unit 24-4, and an accuracy comparison unit 29. The function processing units 22-4k, the data restoration unit 24-4, and the accuracy comparison unit 29 are configured by a controller. The recording unit 23-4 is configured by a memory.

[0119] In the example shown in FIG. 17, the plurality of function processing units 22-4k are shown as a first function processing unit 22-41, a second function processing unit 22-42, a third function processing unit 22-43, . . .

[0120] Each of the plurality of function processing units 22-4k estimates parameters of a plurality of different types of distribution functions that indicate the distribution of traffic data for each predetermined period. The specific process by which the plurality of function processing units 22-4k estimates the parameters of each distribution function is similar to the process of the function processing unit 22 in the first embodiment.

[0121] 17, the first function processing unit 22-41 can estimate the parameters of a distribution function indicating a normal distribution, the second function processing unit 22-42 can estimate the parameters of a distribution function indicating a triangular distribution, and the third function processing unit 22-43 can estimate the parameters of a distribution function indicating a Lorentz distribution. Furthermore, without being limited to this, the k-th function processing unit 22-4k can estimate the parameters of an arbitrary distribution function different from those of the other function processing units.

[0122] The distribution function representing the normal distribution is expressed by the following equation (2).

[0123]

number

[0124] The distribution function representing the triangular distribution is expressed by the following formula (3): In formula (3), a, b, and c are parameters.

[0125]

number

[0126] The distribution function indicating the Lorentz distribution is expressed by the following equation (4): In equation (4), x0 and γ are parameters.

[0127]

number

[0128] Furthermore, each of the plurality of function processing units 22-4k outputs the calculated parameters to the accuracy comparison unit 29.

[0129] The accuracy comparing unit 29 selects the parameters of the distribution functions that have the smallest error (e.g., mean square error) between the traffic distribution indicated by each of the plurality of distribution functions and the traffic distribution of the traffic data acquired by the data acquiring unit 21. Then, the accuracy comparing unit 29 associates the selected parameters with parameter information and records them in the recording unit 23-4. The parameter information is information for identifying the type of distribution function from which the parameters are calculated. The parameter information may be the type of parameter (e.g., standard deviation, mean value), or may be information indicating the type of distribution function from which the parameters are calculated (e.g., normal distribution).

[0130] The recording unit 23-4 records the selected parameters in association with parameter information for identifying the type of distribution function corresponding to the parameters.

[0131] The data restoration unit 24-4 acquires the parameters and parameter information of the distribution function recorded in the recording unit 23-4. Then, the data restoration unit 24-4 generates restored traffic data based on the parameters and the type of distribution function identified by the parameter information. The specific processing by the data restoration unit 24-4 to generate restored traffic data is similar to the processing by the data restoration unit 24 in the first embodiment.

[0132] <Operation of the data recording device> Next, the operation of the data recording device 2-4 according to the fifth embodiment will be described with reference to Fig. 18. Fig. 18 is a sequence diagram showing an example of the operation of the data recording device 2-4 according to the fifth embodiment. The operation of the data recording device 2-4 described with reference to Fig. 18 corresponds to the data recording method executed by the data recording device 2-4 according to the fifth embodiment.

[0133] In step S51, the data acquisition unit 21 acquires traffic data indicating the amount of traffic generated at each predetermined time interval.

[0134] In step S52, the plurality of function processing units 22-4k estimate parameters of a plurality of distribution functions of different types that indicate the distribution of traffic data for each predetermined period.

[0135] In step S53, the accuracy comparing unit 29 selects the parameters of the distribution function that has the smallest error between the traffic distribution indicated by each of the plurality of distribution functions and the traffic distribution of the traffic data acquired by the data acquiring unit 21.

[0136] In step S54, the recording unit 23-4 records the selected parameters in association with parameter information for identifying the type of distribution function corresponding to the parameters.

[0137] The operation of the data recording device 2-4 in the fifth embodiment to generate restored traffic data is similar to the operation of the data recording device 2 in the first embodiment to generate restored traffic data, but the type of distribution function used in this operation is determined by the parameter information recorded in the recording unit 23-4.

[0138] As described above, according to the fifth embodiment, the data recording device 2-4 includes a plurality of function processing units 22-4k that estimate, for each predetermined period, parameters of a plurality of different types of distribution functions that indicate the distribution of traffic data, and an accuracy comparing unit 29 that selects the parameters of the distribution function that has the smallest error between the traffic distribution indicated by each of the plurality of distribution functions and the traffic distribution of the traffic data acquired by the data acquiring unit 21, and the recording unit 23-4 records the selected parameters in association with parameter information for identifying the type of distribution function corresponding to the parameters.

[0139] This allows the data recording device 2-4 to record parameters of a distribution function that reduces the error between the traffic distribution of the traffic data acquired by the data acquiring unit 21 and the traffic data. This allows the data recording device 2-4 to suppress the deviation between the traffic data collected by the traffic collecting device 1 and the restored traffic data. In other words, the data recording device 2-4 can improve the reproducibility of the restored traffic data.

[0140] <<Sixth embodiment>> The overall configuration of the sixth embodiment will be described with reference to Fig. 19. Fig. 19 is a schematic diagram showing an example of a traffic prediction system 100-5 according to the sixth embodiment.

[0141] The traffic prediction system 100-5 includes a traffic collection device 1, a data recording device 2-5, a bandwidth prediction device 3, and a line information storage device 4. The traffic collection device 1 and the line information storage device 4 communicate with the data recording device 2-5 via a communication network. The data recording device 2-5 and the bandwidth prediction device 3 also communicate with each other via a communication network. In the sixth embodiment, the same functional units as those in the first embodiment are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0142] The data recording device 2-5 includes a data acquisition unit 21-5, a function processing unit 22-5, a recording unit 23-5, a data restoration unit 24, a determination unit 25-5, and a function synthesis unit 26-5. The data acquisition unit 21-5 is configured by an input interface. The function processing unit 22-5, the determination unit 25-5, and the function synthesis unit 26-5 are configured by a controller. The recording unit 23-5 is configured by a memory.

[0143] Similar to the data acquisition unit 21-1 of the second embodiment, the data acquisition unit 21-5 acquires traffic data indicating the amount of traffic generated at predetermined times from the traffic collection device 1 via the communication network at any timing. Also, similar to the data acquisition unit 21-1 of the second embodiment, the data acquisition unit 21-5 acquires line information corresponding to the traffic data from the line information storage device 4 via the communication network at any timing. The line information in this embodiment is similar to the line information in the second embodiment.

[0144] The function processing unit 22 estimates parameters of a distribution function that indicates a traffic distribution during a predetermined period that includes a plurality of consecutive predetermined times. Specifically, the function processing unit 22 estimates a first parameter that is a parameter that indicates a distribution function during a first predetermined period, and, in a state in which the first parameter is recorded in the recording unit 23-5, estimates a second parameter that is a parameter that indicates a distribution function during a second predetermined period that follows the first predetermined period.

[0145] Similar to the determination unit 25 of the second embodiment, the determination unit 25-5 determines whether the line information (e.g., contracted bandwidth) for the second predetermined period is the same as the line information for the first predetermined period. At this time, the determination unit 25-5 acquires the line information for the first predetermined period from the recording unit 23-5 and acquires the line information for the second predetermined period from the function processing unit 22, and performs the above-mentioned determination. Note that if multiple lines are contracted for the predetermined period, the determination unit 25-5 determines whether the line information for each of the multiple lines is the same.

[0146] The determination unit 25 in the second embodiment determines whether the difference between the first parameter and the second parameter is equal to or less than the first threshold value. However, the determination unit 25-5 in the sixth embodiment does not execute the process of determining whether the difference is equal to or less than the first threshold value.

[0147] When it is determined that the line information (e.g., contract bandwidth) for the second predetermined period is the same as the line information for the first predetermined period, the function synthesis unit 26-5 calculates a weighted synthesis parameter based on a value obtained by weighting the first parameter and the second parameter. Specifically, the function synthesis unit 26-5 may calculate the average value of the weighted values ​​of the first parameter and the second parameter as the weighted synthesis parameter.

[0148] Furthermore, the function synthesis unit 26-5 may calculate a weighted synthesis parameter by assigning a larger weight to a parameter of a distribution function that indicates a traffic distribution in the newer of the first and second predetermined periods. Furthermore, the weight to be assigned to the parameter may be determined in advance by an operator for each predetermined period (for example, one day).

[0149] The function synthesis unit 26-5 causes the weighting synthesis parameters to be recorded in the recording unit 23-5.

[0150] If the determining unit 25-5 determines that the channel information is different, the function combining unit 26-5 does not calculate the weighting combining parameter.

[0151] The recording unit 23-5 records the weighted synthesis parameter calculated by the function synthesis unit 26-5. When the determination unit 25-5 determines that the line information for the second predetermined period is not the same as the line information for the first predetermined period, the recording unit 23-5 records the first parameter, the second parameter, and the corresponding contract bandwidth.

[0152] <Operation of the data recording device> Next, the operation of the data recording device 2-5 according to the sixth embodiment will be described with reference to Fig. 20. Fig. 20 is a sequence diagram showing an example of the operation of the data recording device 2-5 according to the sixth embodiment. The operation of the data recording device 2-5 described with reference to Fig. 20 corresponds to the data recording method executed by the data recording device 2-5 according to the sixth embodiment. In the example shown in Fig. 20, at the start of the operation, the function processing unit 22-5 calculates a first parameter based on traffic data for a first predetermined period, and the recording unit 23-5 records the first parameter. In addition, line information for the first predetermined period is acquired.

[0153] In step S61, the data acquisition unit 21-1 acquires traffic data from the traffic collection device 1 via the communication network.

[0154] In step S62, the data acquisition unit 21-1 acquires line information for each line subscribed to by the user from the line information storage device 4 via the communication network.

[0155] In step S63, the function processing unit 22-1 estimates a second parameter based on the traffic data for a second predetermined period.

[0156] In step S64, the determination unit 25 acquires from the recording unit 23 the first parameter and line information for the first predetermined period.

[0157] In step S65, the determination unit 25 determines whether the line information in the second predetermined period is the same as the line information in the first predetermined period.

[0158] If the determination unit 25 determines in step S65 that the line information in the first predetermined period is not the same as the line information in the second predetermined period, the recording unit 23-1 further records the second parameter in step S66.

[0159] If it is determined in step S65 that the line information for the first predetermined period is the same as the line information for the second predetermined period, in step S67, the function synthesis unit 26 calculates a weighted synthesis parameter based on a value obtained by weighting the first parameter and the second parameter.

[0160] In step S68, the recording unit 23-1 records the weighted synthesis parameter. Specifically, the recording unit 23-1 updates the first parameter recorded in the recording unit 23-1 to the weighted synthesis parameter.

[0161] The operation of the data recording device 2-5 in the sixth embodiment to generate restored traffic data is similar to the operation of the data recording device 2 in the first embodiment to generate restored traffic data.

[0162] As described above, according to the sixth embodiment, the data recording device 2-5 further includes a determination unit 25-5 and a function synthesis unit 26-5, the data acquisition unit 21-5 acquires line information corresponding to traffic data, the function processing unit 22 estimates a first parameter which is a parameter indicating a distribution function for a first predetermined period, and, while the first parameter is recorded in the recording unit 23-5, estimates a second parameter which is a parameter indicating a distribution function for a second predetermined period after the first predetermined period, the determination unit 25-5 determines whether the line information for the second predetermined period is identical to the line information for the first predetermined period, and, if it is determined that the line information for the second predetermined period is identical to the line information for the first predetermined period, the function synthesis unit 26-5 calculates a weighted synthesis parameter based on a value obtained by weighting the first parameter and the second parameter.

[0163] As a result, the data recording device 2-5 records composite parameters, which are parameters of a function that indicates the distribution of traffic data over multiple predetermined periods, and therefore can further reduce the amount of data to be recorded compared to a configuration that records parameters for each predetermined period. Furthermore, unlike the data recording device 2-1 of the second embodiment, the amount of data to be recorded can be reduced even when there is a large deviation in the traffic distribution over multiple predetermined periods, i.e., when the traffic distribution changes over time. Furthermore, the data recording device 2-5 can generate restored traffic data that strongly reflects the trends of newly collected traffic data by assigning a larger weight to the parameter of the distribution function that indicates the traffic distribution over a newer period among the multiple predetermined periods.

[0164] <Program> The above-described data recording device 2, data recording device 2-1, data recording device 2-2, data recording device 2-3, data recording device 2-4, and data recording device 2-5 can be realized by a computer 101. Furthermore, a program for causing the above-described data recording device 2, data recording device 2-1, data recording device 2-2, data recording device 2-3, data recording device 2-4, and data recording device 2-5 to function may be provided. Furthermore, the program may be stored on a storage medium or provided via a network. FIG. 21 is a block diagram showing a schematic configuration of a computer 101 functioning as the data recording device 2. The computers functioning as the data recording device 2-1, data recording device 2-2, data recording device 2-3, data recording device 2-4, and data recording device 2-5 may also be configured similarly to the computer 101. Here, the computer 101 may be a general-purpose computer, a dedicated computer, a workstation, a PC (Personal Computer), an electronic notepad, or the like. The program instructions may be program code, code segments, or the like for executing necessary tasks.

[0165] 21, the computer 101 includes a processor 110, a ROM (Read Only Memory) 120, a RAM (Random Access Memory) 130, a storage 140, an input unit 150, an output unit 160, and a communication interface (I / F) 170. Each component is connected to each other via a bus 180 so as to be able to communicate with each other. The processor 110 is specifically a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an SoC (System on a Chip), or the like, and may be configured by a plurality of processors of the same type or different types.

[0166] The processor 110 controls each component and performs various arithmetic processing. That is, the processor 110 reads a program from the ROM 120 or the storage 140 and executes the program using the RAM 130 as a work area. The processor 110 controls each component and performs various arithmetic processing in accordance with the program stored in the ROM 120 or the storage 140. In the above-described embodiment, the program according to the present disclosure is stored in the ROM 120 or the storage 140.

[0167] The program may be stored in a storage medium readable by the computer 101. Using such a storage medium, the program can be installed in the computer 101. Here, the storage medium on which the program is stored may be a non-transitory storage medium. The non-transitory storage medium is not particularly limited, and may be, for example, a CD-ROM, a DVD-ROM, or a USB (Universal Serial Bus) memory. Furthermore, the program may be downloaded from an external device via a network.

[0168] The ROM 120 stores various programs and various data. The RAM 130 temporarily stores programs or data as a working area. The storage 140 is configured with an HDD (Hard Disk Drive) or SSD (Solid State Drive) and stores various programs including the operating system and various data.

[0169] The input unit 150 includes one or more input interfaces that receive input operations from a user and acquire information based on the user operations. For example, the input unit 150 is, but is not limited to, a pointing device, a keyboard, a mouse, etc.

[0170] The output unit 160 includes one or more output interfaces that output information. For example, the output unit 160 is a display that outputs information as a video or a speaker that outputs information as an audio, but is not limited to these. Note that if the output unit 160 is a touch panel display, it also functions as the input unit 150.

[0171] The communication interface (I / F) 170 is an interface for communicating with an external device.

[0172] The following additional notes are provided regarding the above-described embodiments. [Additional note 1] The system includes an input interface for acquiring traffic data indicating the amount of traffic generated at each predetermined time, a controller, and a memory; The controller Estimating parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times; The memory stores the parameters; The controller generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as restored traffic data. [Additional note 2] the input interface acquires line information corresponding to the traffic data; The controller estimating a first parameter that is a parameter indicating the distribution function during a first predetermined period, and estimating a second parameter that is a parameter indicating the distribution function during a second predetermined period after the first predetermined period while the first parameter is recorded in the memory; determining whether the line information during the second predetermined period is the same as the line information during the first predetermined period, and if it is determined that the line information during the second predetermined period is the same as the line information during the first predetermined period, determining whether a difference between the first parameter and the second parameter is equal to or less than a first threshold; If it is determined that the difference is equal to or smaller than a first threshold, calculating a composite parameter based on the first parameter and the second parameter; 2. The data recording device according to claim 1, wherein the memory records the synthesis parameters. [Additional note 3] the controller extracts deviant traffic data from the traffic data acquired by the input interface, the deviant traffic data being traffic data that deviates from a traffic distribution indicated by a distribution function having the estimated parameters; The memory further records the off-traffic data; The data recording device according to claim 1 or 2, wherein the controller generates, as restored traffic data, traffic data that follows a traffic distribution indicated by the distribution function having the parameters and the deviating traffic data. [Additional note 4] The controller For each predetermined period, a superposition distribution function including n (n is an integer of 1 or more) distribution functions is estimated; determining whether an error between the traffic distribution indicated by the superposition distribution function and the traffic distribution of the traffic data acquired by the input interface is equal to or less than a second threshold; the memory records parameters of the superposition distribution function when it is determined that the error is equal to or less than the second threshold; 3. The data recording device according to claim 1, wherein, when it is determined that the error is greater than the second threshold, the controller sets n=n+1 and again estimates parameters of a superposition distribution function including n distribution functions. [Additional note 5] The controller estimating parameters of a plurality of distribution functions of different types that indicate a distribution of the traffic data for each predetermined period; selecting parameters of a distribution function that has the smallest error between the traffic distribution indicated by each of the plurality of distribution functions and the traffic distribution of the traffic data acquired by the input interface; 3. The data recording device according to claim 1, wherein the memory records the selected parameters in association with parameter information for identifying the type of distribution function corresponding to the parameters. [Additional note 6] the input interface acquires line information corresponding to the traffic data; The controller estimating a first parameter that is a parameter indicating the distribution function during a first predetermined period, and estimating a second parameter that is a parameter indicating the distribution function during a second predetermined period after the first predetermined period, while the first parameter is recorded in the recording unit; determining whether the line information during the second predetermined period is the same as the line information during the first predetermined period; The data recording device described in appendix 1, wherein if it is determined that the line information for the second specified period is the same as the line information for the first specified period, a weighted composite parameter is calculated based on a weighted value of the first parameter and the second parameter. [Additional note 7] Acquire traffic data indicating the amount of traffic generated at each predetermined time interval; estimating parameters of a distribution function that indicates a distribution of the traffic data during a predetermined period that includes a plurality of consecutive predetermined times; recording said parameters; generating traffic data according to the distribution function having the parameters as restored traffic data based on the parameters; Data recording method. [Additional note 8] A non-transitory storage medium storing a program executable by a computer, the program causing the computer to operate as the data recording device described in appendix 1 or 2.

[0173] All publications, patent applications, and technologies mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and technology was specifically and individually indicated to be incorporated by reference.

[0174] Although the above-described embodiments have been described as typical examples, it will be apparent to those skilled in the art that many modifications and substitutions can be made within the spirit and scope of the present disclosure. Therefore, the present invention should not be construed as being limited by the above-described embodiments, and various modifications or alterations are possible without departing from the scope of the claims. [Explanation of symbols]

[0175] 1 Traffic collection device 2, 2-1, 2-2, 2-3, 2-4, 2-5 Data recording device 3 Bandwidth Prediction Device 4 Line information storage device 21, 21-1, 21-5 Data acquisition section 22, 22-1, 22-3, 22-4k, 22-5 Function processing section 23, 23-1, 23-2, 23-4, 23-5 Recording section 24, 24-2, 24-4 Data recovery section 25, 25-5 Judgment section 26, 26-5 Function composition section 27 Outlier extraction unit 28 Overlay judgment unit 29 Accuracy Comparison Section 100, 100-1, 100-2, 100-3, 100-4, 100-5 Traffic Prediction System 101 Computer 110 processors 120 ROM 130 RAM 140 Storage 150 Input section 160 Output section 170 Communication Interface 180 Bus

Claims

1. a data acquisition unit that acquires traffic data indicating the amount of traffic generated at each predetermined time; a function processing unit that estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data during a predetermined period that includes a plurality of consecutive predetermined times; a recording unit that records the parameters; a data recovery unit that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as recovered traffic data; a determination unit, a function synthesis unit, Equipped with the data acquisition unit acquires line information corresponding to the traffic data; the function processing unit estimates a first parameter that is a parameter indicating the distribution function for a first predetermined period, and, in a state in which the first parameter is recorded in the recording unit, estimates a second parameter that is a parameter indicating the distribution function for a second predetermined period after the first predetermined period; the determination unit determines whether or not line information during the second predetermined period is identical to line information during the first predetermined period, and when it is determined that the line information during the second predetermined period is identical to the line information during the first predetermined period, determines whether or not a difference between the first parameter and the second parameter is equal to or less than a first threshold; the function synthesis unit calculates a synthesis parameter based on the first parameter and the second parameter when it is determined that the difference is equal to or smaller than a first threshold value; The recording unit is a data recording device that records the synthesis parameters.

2. A data acquisition unit that acquires traffic data indicating the amount of traffic generated at each predetermined time; a function processing unit that estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data during a predetermined period that includes a plurality of consecutive predetermined times; a recording unit that records the parameters; a data recovery unit that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as recovered traffic data; an outlier extraction unit that extracts outlier traffic data, which is traffic data that deviates from a traffic distribution indicated by a distribution function having parameters estimated by the function processing unit, from the traffic data acquired by the data acquisition unit; Equipped with The recording unit further records the off-traffic data, The data recovery unit is a data recording device that generates recovered traffic data from traffic data that follows a traffic distribution indicated by the distribution function having the parameters and the deviating traffic data.

3. A data acquisition unit that acquires traffic data indicating the amount of traffic generated at each predetermined time; a function processing unit that estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data during a predetermined period that includes a plurality of consecutive predetermined times; a recording unit that records the parameters; a data recovery unit that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as recovered traffic data; an overlay determination unit; Equipped with the function processing unit estimates a superposition distribution function including n (n is an integer equal to or greater than 1) distribution functions for each predetermined period; the overlap determination unit determines whether an error between the traffic distribution indicated by the overlap distribution function and the traffic distribution of the traffic data acquired by the data acquisition unit is equal to or less than a second threshold; the recording unit records parameters of the superposition distribution function when it is determined that the error is equal to or less than the second threshold; When it is determined that the error is greater than the second threshold, the function processing unit sets n=n+1 and again estimates parameters of a superposition distribution function including n distribution functions.

4. a plurality of said function processing units that estimate parameters of a plurality of distribution functions of different types that indicate a distribution of said traffic data for each predetermined period; an accuracy comparison unit that selects a parameter of a distribution function that has the smallest error between a traffic distribution indicated by each of the plurality of distribution functions and a traffic distribution of the traffic data acquired by the data acquisition unit; 2. The data recording device according to claim 1, wherein the recording section records the selected parameter in association with parameter information for identifying a type of distribution function corresponding to the parameter.

5. A data acquisition unit that acquires traffic data indicating the amount of traffic generated at each predetermined time; a function processing unit that estimates parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data during a predetermined period that includes a plurality of consecutive predetermined times; a recording unit that records the parameters; a data recovery unit that generates, based on the parameters, traffic data that conforms to the distribution function having the parameters as recovered traffic data; a determination unit, a function synthesis unit, Equipped with the data acquisition unit acquires line information corresponding to the traffic data; the function processing unit estimates a first parameter that is a parameter indicating the distribution function for a first predetermined period, and, in a state in which the first parameter is recorded in the recording unit, estimates a second parameter that is a parameter indicating the distribution function for a second predetermined period after the first predetermined period; the determination unit determines whether the line information for the second predetermined period is the same as the line information for the first predetermined period; A data recording device in which, when it is determined that the line information for the second specified period is identical to the line information for the first specified period, the function synthesis unit calculates a weighted synthesis parameter based on a value obtained by weighting the first parameter and the second parameter.

6. acquiring traffic data indicating the amount of traffic generated at each predetermined time; estimating parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times; recording said parameters; generating traffic data according to the distribution function having the parameters as restored traffic data based on the parameters; A determination step, a function synthesis step, Including, the acquiring step includes acquiring line information corresponding to the traffic data; the estimating step includes estimating a first parameter that is a parameter indicating the distribution function in a first predetermined period, and estimating a second parameter that is a parameter indicating the distribution function in a second predetermined period after the first predetermined period while the first parameter is being recorded; The determining step determines whether or not line information during the second predetermined period is identical to line information during the first predetermined period, and if it is determined that the line information during the second predetermined period is identical to the line information during the first predetermined period, determines whether or not a difference between the first parameter and the second parameter is equal to or less than a first threshold value; the function synthesis step calculates a synthesis parameter based on the first parameter and the second parameter when it is determined that the difference is equal to or smaller than a first threshold value; The recording step is a data recording method in which the synthesis parameters are recorded.

7. A step of acquiring traffic data indicating the amount of traffic generated at each predetermined time; estimating parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times; recording said parameters; generating traffic data according to the distribution function having the parameters as restored traffic data based on the parameters; extracting deviant traffic data from the traffic data acquired in the acquiring step, the deviant traffic data being traffic data that deviates from the traffic distribution indicated by the distribution function having the parameters estimated in the estimating step; Including, The recording step further includes recording the off-traffic data; The generating step generates, as restored traffic data, traffic data that follows a traffic distribution represented by the distribution function having the parameters and the deviating traffic data.

8. A step of acquiring traffic data indicating the amount of traffic generated at each predetermined time; estimating parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times; recording said parameters; generating traffic data according to the distribution function having the parameters as restored traffic data based on the parameters; a superposition step; Including, the estimating step estimates a superposition distribution function including n (n is an integer of 1 or more) distribution functions for each predetermined period; the superposition step determines whether an error between the traffic distribution indicated by the superposition distribution function and the traffic distribution of the traffic data acquired in the acquiring step is equal to or less than a second threshold; the recording step records parameters of the superposition distribution function when it is determined that the error is equal to or less than the second threshold; The data recording method, wherein the estimating step, if it is determined that the error is greater than the second threshold, sets n=n+1 and again estimates parameters of a superposition distribution function including n distribution functions.

9. A step of acquiring traffic data indicating the amount of traffic generated at each predetermined time; estimating parameters of a distribution function that indicates a traffic distribution, which is a distribution of the traffic data over a predetermined period that includes a plurality of consecutive predetermined times; recording said parameters; generating traffic data according to the distribution function having the parameters as restored traffic data based on the parameters; A determination step, a function synthesis step, Including, the acquiring step includes acquiring line information corresponding to the traffic data; the estimating step includes estimating a first parameter that is a parameter indicating the distribution function in a first predetermined period, and estimating a second parameter that is a parameter indicating the distribution function in a second predetermined period after the first predetermined period while the first parameter is being recorded; the determining step determines whether the line information in the second predetermined period is the same as the line information in the first predetermined period; A data recording method in which the function synthesis step calculates a weighted synthesis parameter based on a value obtained by weighting the first parameter and the second parameter when it is determined that the line information for the second specified period is the same as the line information for the first specified period.

10. A program for causing a computer to function as the data recording device according to any one of claims 1 to 3 or 5.

Images