JP2004048276A - Method and apparatus for estimating peak traffic, program for estimating peak traffic, and storage medium having the program for estimating peak traffic recorded thereon - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To estimate peak traffic with high reliability by taking account of stochastic fluctuations. <P>SOLUTION: A method for estimating peak traffic includes the steps of: estimating a probability distribution which data whose value exceeds a prescribed reference value follows among received measurement data of a network of a noted communication service and selecting the right end of the probability distribution to be the estimate value of the peak traffic; dividing the received measurement data of the network of the noted communication service into a plurality of groups; acquiring data taking a maximum value in each of the divided groups; estimating the probability distribution which the acquired maximum value data follow; and selecting the right end of the probability distribution to be the estimated value of the peak traffic. <P>COPYRIGHT: (C)2004,JPO

Description

とおく。ここで、ξ、μ、σは、パラメータであり、
【００２９】
【数２】

は、一般パレート分布と呼ばれる。
【００３０】
ところで、極値理論に基づけば、適当な仮定の下で基準値ｘｋ　を上側に超過するデータの超過幅の分布
Ｐｒ　｛Ｘ−ｘｋ　＜ｘ｜Ｘ＞ｘｋ　｝
は、
【００３１】
【数３】

へ収束する。
【００３２】
本実施の形態においては、極値理論の仮定が成り立ち、基準値ｘｋ　を超過するデータの超過幅が一般パレート分布に従うとする。超過幅の分布がわかれば、超過幅に基準値ｘｋ　を加えることで、基準値ｘｋ　を超えるデータが従う確率分布を推定できる。
【００３３】
ここで、基準値を上側に超過するデータの超過幅の分布を推定する確率分布推定部１２０（即ち、一般パレート分布のパラメータを推定する手段）について説明する。まず、基準値ｘｋ　を超過するデータの超過幅の平均ｅｋ　、分散ｖｋ　、変動係数の２乗ｃｋ　をそれぞれ、以下のように求める。
【００３４】
【数４】

そして、一般パレート分布のパラメータξｋ　，σｋ　の推定値を以下ように求める（ステップ１０３）。
【００３５】
【数５】

この
【００３６】
【数６】

を用いることで、基準値を上側に超過する幅の確率分布の分布関数は、
【００３７】
【数７】

と推定される（ステップ１０４）。
【００３８】
また、
【００３９】
【数８】

を次のようにおく。
【００４０】
【数９】

以上により、ｘ＞ｘｋ　について、基準値ｘｋ　を超えるデータが従う確率分布の分布関数は、
【００４１】
【数１０】

と推定できる（ステップ１０５）。
【００４２】
最後に、ピークトラヒック推定部１３０について説明する。
【００４３】
もし、
【００４４】
【数１１】

が「負」であれば、上記確率分布の右端は有界であり、それは、
【００４５】
【数１２】

と求められる。従って、これをピークトラヒックの推定値とする（ステップ１０６）。一方、
【００４６】
【数１３】

が「正」であれば、上記確率分布の右端は有界とならないので、従来の技術などの他の手法を用いてピークトラヒックを推定する（ステップ１０７）。
【００４７】
［第２の実施の形態］
図５は、本発明の第２の実施の形態におけるピークトラヒック推定装置の構成を示す。
【００４８】
同図に示すピークトラヒック推定装置は、測定データ分割部２１０、最大値データ取得部２２０、最悪値データ分布推定部２３０、ピークトラヒック推定部２４０から構成される。
【００４９】
測定データ分割部２１０は、着目する通信サービスのネットワークに関する測定データを複数のグループに分割する。
【００５０】
最大値データ取得部２２０は、測定データ分割部２１０で分割されたグループから、そのグループの中で最大値をとるデータを取得する。
【００５１】
最悪値データ分布推定部２３０は、最大値データ取得部２２０で取得された最大値データが従う確率分布を推定する。
【００５２】
ピークトラヒック推定部２４０は、最悪値データ分布推定部２３０で推定された確率分布の右端をピークトラヒックの推定値とする。
【００５３】
次に、上記の構成における動作を説明する。
【００５４】
図６は、本発明の第２の実施の形態における動作のフローチャートである。
【００５５】
以下では、ｔ＝ｍｎとする。このとき、時系列データは、
【００５６】
【数１４】

のように表すことができる。
【００５７】
まず、測定データ分割部２１０により、｛ｚ１　，…，ｚｍｎ｝をｍ個のグループに分割し、ｉ番目のグループを｛ｚ（ｉ−１）ｎ＋１，ｚ（ｉ−１）ｎ＋２，…，ｚｉｎ｝とおく（１≦ｉ≦ｍ）（ステップ２０１）。
【００５８】
次に、最大値データ取得部２２０により、以下のように最大値データｘｉ　を取得する（１≦ｉ≦ｍ）（ステップ２０２）。
【００５９】
ｘｉ　＝ｍａｘ　｛ｚ（ｉ−１）ｎ＋１，ｚ（ｉ−１）ｎ＋２，…，ｚｉｎ｝
ところで、極値理論に基づけば、適当な仮定の下で最大値の分布は、一般極値分布へ収束する。一般極値分布とは、分布関数
【００６０】
【数１５】

が次のように、与えられる確率分布である。ここで、ξ，μ，σはパラメータである。
【００６１】
【数１６】

本実施の形態においては、極値理論の仮定が成り立ち、最大値データ｛ｘ１　，…，ｘｍ　｝が一般極値分布に従うとする。
【００６２】
ここで、最大値の確率分布を推定する最悪値データ分布推定部２３０について説明する。まず、最大値データの平均ｅ，分散ｖ、歪度ｓをそれぞれ以下のように求める（ステップ２０３）。
【００６３】
【数１７】

そして、一般極値分布のパラメータをξ，σ，μの推定値を以下のように求める（ステップ２０４）。
【００６４】
ステップ２０４−１）次の方程式を解き、ξの推定値
【００６５】
【数１８】

を決定する。但し、
【００６６】
【数１９】

である。また、Γ（・）はガンマ関数である。
【００６７】
【数２０】

ステップ２０４−２）ステップ２０４−１の結果を用い、以下のように、σｊ　の推定値
【００６８】
【数２１】

を決定する。
【００６９】
【数２２】

ステップ２０４−３）ステップ２０４−１及びステップ２０４−２の結果を用い、以下のように、μの推定値
【００７０】
【数２３】

を決定する。
【００７１】
【数２４】

求められた
【００７２】
【数２５】

を用いることで、最大値データの従う確率分布の分布関数は、
【００７３】
【数２６】

と推定される（ステップ２０５）。
【００７４】
最後に、ピークトラヒック推定部２４０について説明する。もし、
【００７５】
【数２７】

が「負」であれば、上記確率分布の右端は有界であり、それは、
【００７６】
【数２８】

と求められる。従って、これがピークトラヒックの推定値となる（ステップ２０６）。一方、
【００７７】
【数２９】

が「正」であれば、上記確率分布の右端は有界とならないので、従来の技術などの他の手法を用いてピークトラヒックを推定する（ステップ２０７）。
【００７８】
なお、上記の実施の形態における図４及び図６の動作をプログラムとして構築し、ピークトラヒック推定装置として利用されるコンピュータにインストールする、または、ネットワークを介して流通させることも可能である。
【００７９】
また、構築されたプログラムをピークトラヒック推定装置として利用されるコンピュータに接続されるハードディスク装置や、フレキシブルディスク、ＣＤ−ＲＯＭ等の可搬記憶媒体に格納しておき、本発明を実施する際にインストールするとも可能である。
【００８０】
なお、本発明は、上記の実施の形態に限定されることなく、特許請求の範囲内において、種々変更・応用が可能である。
【００８１】
【発明の効果】
上述のように、本発明によれば、着目する通信サービスのネットワークに関する測定データが得られた際に、測定データ全体が従う確率分布を同定せずに、通信品質指標が正常値から大きく離れた値、あるいは、最大値を取る場合における確率的挙動のみを推定することで、従来よりも信頼性の高いピークトラヒック推定値を効率的に決定することができる。
【図面の簡単な説明】
【図１】本発明の原理を説明するための図である。
【図２】本発明の原理構成図である。
【図３】本発明の第１の実施の形態におけるピークトラヒック推定装置の構成図である。
【図４】本発明の第１の実施の形態における動作のフローチャートである。
【図５】本発明の第２の実施の形態におけるピークトラヒック推定装置の構成図である。
【図６】本発明の第２の実施の形態における動作のフローチャートである。
【符号の説明】
１００　ピークトラヒック推定装置
１１０　基準値決定部
１２０　確率分布推定手段、確率分布推定部
１３０　ピークトラヒック推定手段、ピークトラヒック推定部
１４０　入力部
１５０　出力部
２００　ピークトラヒック推定装置
２１０　測定データ分割部
２２０　最大値データ取得部
２３０　最悪値データ分布推定部
２４０　ピークトラヒック推定部
２５０　入力部
２６０　出力部[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a peak traffic estimating method and apparatus, a peak traffic estimating program, and a storage medium storing the peak traffic estimating program. And method for estimating peak traffic, a peak traffic estimating program, and a storage medium storing the peak traffic estimating program.
[0002]
[Prior art]
As a conventional method for estimating peak traffic, there is a method in which a maximum value of a measured value related to a communication quality index is used as an estimated value of peak traffic.
[0003]
[Problems to be solved by the invention]
It can be considered that the measured value of the communication quality index always fluctuates stochastically. However, in the above-mentioned conventional technology, there is a problem that the probability of the estimated value of the peak traffic being lower than the true peak traffic becomes strong because this stochastic fluctuation is not considered.
[0004]
By the way, when estimating the peak traffic, it is more important to accurately grasp the probabilistic behavior when the index takes a value far away from the index or when the index takes the maximum value, rather than when the index takes an average value. . Therefore, in order to estimate the peak traffic, it is sufficient to accurately estimate only the stochastic behavior when the index exceeds a certain reference value or when the index takes the maximum value. In many cases, it is not necessary to estimate accurately.
[0005]
The present invention has been made in view of the above points, and in a network of a communication service of interest, when measurement data related to a communication quality index is obtained, without fixing a probability distribution followed by the entire measurement data, an index is used. A peak traffic estimation method and apparatus for estimating peak traffic with higher reliability than before by accurately estimating only a stochastic behavior when a value exceeding a certain reference value or a maximum value is taken, and a peak traffic It is an object to provide a storage medium storing an estimation program and a peak traffic estimation program.
[0006]
[Means for Solving the Problems]
FIG. 1 is a diagram for explaining the principle of the present invention.
[0007]
The present invention provides a peak traffic estimating method for estimating peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest.
Estimate the probability distribution according to the data whose data value exceeds a predetermined reference value among the measurement data relating to the network of the communication service of interest input by the input means (step 1);
The right end of the probability distribution is output as an estimated value of the peak traffic (step 2). The present invention provides a peak traffic estimating method for estimating peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest.
The measurement data regarding the network of the communication service of interest input by the input unit is divided into a plurality of groups,
From each of the divided groups, obtain the data that takes the maximum value in the group,
Estimate the probability distribution that the acquired maximum value data follows,
The right end of the probability distribution is output as an estimated value of the peak traffic.
[0008]
FIG. 2 is a diagram illustrating the principle of the present invention.
[0009]
The present invention is a peak traffic estimating apparatus for estimating peak traffic serving as a guideline when designing, managing and controlling a network of a communication service of interest,
Probability distribution estimating means 120 for estimating a probability distribution followed by data whose data value exceeds a predetermined reference value among measurement data relating to the network of the communication service of interest inputted by the input means,
A peak traffic estimating unit 130 that outputs the right end of the probability distribution estimated by the probability distribution estimating unit 120 as an estimated value of the peak traffic.
[0010]
The present invention is a peak traffic estimating apparatus for estimating peak traffic serving as a guideline when designing, managing and controlling a network of a communication service of interest,
Measurement data division means for dividing the measurement data regarding the network of the communication service of interest inputted by the input means into a plurality of groups,
From each group divided by the measurement data dividing unit, a maximum value data acquiring unit for acquiring data having a maximum value in the group,
A probability distribution estimating means for estimating a probability distribution according to the acquired maximum value data;
Peak traffic estimating means for outputting the right end of the probability distribution as an estimated value of the peak traffic.
[0011]
The present invention is a peak traffic estimation program for estimating peak traffic that serves as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data regarding a network of a communication service of interest;
Probability distribution estimation step of estimating a probability distribution followed by data whose data value exceeds a predetermined reference value among the measurement data,
A peak traffic estimation step in which the right end of the probability distribution estimated by the probability distribution estimation step is an estimated value of the peak traffic;
Outputting an estimated value of peak traffic.
[0012]
The present invention is a peak traffic estimation program for estimating peak traffic that serves as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data regarding a network of a communication service of interest;
A measurement data dividing step of dividing the measurement data into a plurality of groups;
From each group divided by the measurement data dividing means, a maximum value data obtaining step of obtaining data having a maximum value in the group,
A probability distribution estimating step of estimating a probability distribution according to the obtained maximum value data;
A peak traffic estimation step in which the right end of the probability distribution is an estimated value of the peak traffic,
Outputting an estimated value of peak traffic.
[0013]
The present invention is a storage medium storing a peak traffic estimation program for estimating peak traffic that serves as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data regarding a network of a communication service of interest;
Probability distribution estimation step of estimating a probability distribution followed by data whose data value exceeds a predetermined reference value among the measurement data,
A peak traffic estimation step in which the right end of the probability distribution estimated by the probability distribution estimation step is an estimated value of the peak traffic;
Outputting the estimated value of the peak traffic.
[0014]
The present invention is a storage medium storing a peak traffic estimation program for estimating peak traffic that serves as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data regarding a network of a communication service of interest;
A measurement data dividing step of dividing the measurement data into a plurality of groups;
From each group divided by the measurement data dividing means, a maximum value data obtaining step of obtaining data having a maximum value in the group,
A probability distribution estimating step of estimating a probability distribution according to the obtained maximum value data;
A peak traffic estimation step in which the right end of the probability distribution is an estimated value of the peak traffic,
Outputting the estimated value of the peak traffic.
[0015]
In the conventional technology, the estimated value of the peak traffic is likely to be lower than the true peak traffic because the stochastic fluctuation of the index is not taken into account, whereas the present invention relates to the network of the communication service of interest. When the measurement data is obtained, the probability distribution that the entire measurement data follows is not fixed, and only the stochastic behavior when the communication quality index exceeds a certain reference value or takes the maximum value is accurately estimated. As a result, it is possible to efficiently obtain an estimated value of peak traffic that is more reliable than in the past.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 3 shows a configuration of the peak traffic estimation device according to the first embodiment of the present invention.
[0017]
The peak traffic estimation device shown in FIG. 1 includes a reference value determination unit 110, a probability distribution estimation unit 120, and a peak traffic estimation unit 130.
[0018]
The reference value determination unit 110 determines a reference value from the input measurement data.
[0019]
The probability distribution estimating unit 120 estimates a probability distribution according to data exceeding a certain reference value among data relating to the communication quality index measured in the network of the communication service of interest.
[0020]
The peak traffic estimating unit 130 sets the right end of the probability distribution obtained by the probability distribution estimating unit 120 as an estimated value of the peak traffic.
[0021]
Next, the operation of the above configuration will be described.
[0022]
FIG. 4 is a flowchart of the operation according to the first embodiment of the present invention.
[0023]
Hereinafter, as an example, description will be made using the line usage rate.
[0024]
Consider a network with an arbitrary topology, and focus on a certain link in the network. Then, t pieces of time-series data at equal time intervals relating to the line usage rate of this link are obtained, and the data at time i is set as zi (i = 1,..., T).
[0025]
Here, it is assumed that t = n. For simplicity, {z1, z2,..., Zn} are rearranged in ascending order, and are rewritten as {x1, x2,..., Xn} (xi <xi + 1) (step 101).
[0026]
First, the reference value determination unit 110 determines a certain reference value xk (1 ≦ k ≦ n−2) (step 102).
[0027]
Also,
[0028]
(Equation 1)

far. Here, ξ, μ, and σ are parameters,
[0029]
(Equation 2)

Is called a general Pareto distribution.
[0030]
By the way, based on the extreme value theory, the distribution Pr {X-xk <x | X> xk} of the excess width of the data exceeding the reference value xk upwards under appropriate assumptions.
Is
[0031]
[Equation 3]

Converges to
[0032]
In the present embodiment, it is assumed that the assumption of the extreme value theory holds, and the excess width of data exceeding the reference value xk follows a general Pareto distribution. If the distribution of the excess width is known, a probability distribution followed by data exceeding the reference value xk can be estimated by adding the reference value xk to the excess width.
[0033]
Here, the probability distribution estimating unit 120 for estimating the distribution of the excess width of the data exceeding the reference value upward (that is, means for estimating the parameters of the general Pareto distribution) will be described. First, the average ek, the variance vk, and the square of the variation coefficient ck of the excess width of the data exceeding the reference value xk are obtained as follows.
[0034]
(Equation 4)

Then, estimated values of the parameters ξk, σk of the general Pareto distribution are obtained as follows (step 103).
[0035]
(Equation 5)

[0036]
(Equation 6)

By using, the distribution function of the probability distribution of the width exceeding the reference value above,
[0037]
(Equation 7)

Is estimated (step 104).
[0038]
Also,
[0039]
(Equation 8)

As follows.
[0040]
(Equation 9)

From the above, for x> xk, the distribution function of the probability distribution followed by data exceeding the reference value xk is:
[0041]
(Equation 10)

(Step 105).
[0042]
Finally, the peak traffic estimation unit 130 will be described.
[0043]
if,
[0044]
[Equation 11]

Is negative, the right end of the probability distribution is bounded,
[0045]
(Equation 12)

Is required. Therefore, this is used as an estimated value of the peak traffic (step 106). on the other hand,
[0046]
(Equation 13)

Is positive, the right end of the probability distribution is not bounded, and the peak traffic is estimated using another technique such as a conventional technique (step 107).
[0047]
[Second embodiment]
FIG. 5 shows a configuration of a peak traffic estimation device according to the second embodiment of the present invention.
[0048]
The peak traffic estimating device shown in FIG. 1 includes a measurement data dividing unit 210, a maximum value data acquiring unit 220, a worst value data distribution estimating unit 230, and a peak traffic estimating unit 240.
[0049]
The measurement data dividing unit 210 divides measurement data relating to a communication service network of interest into a plurality of groups.
[0050]
The maximum value data acquisition unit 220 acquires, from the group divided by the measurement data division unit 210, data having the maximum value in the group.
[0051]
The worst value data distribution estimating unit 230 estimates a probability distribution followed by the maximum value data acquired by the maximum value data acquiring unit 220.
[0052]
The peak traffic estimating unit 240 sets the right end of the probability distribution estimated by the worst value data distribution estimating unit 230 as the estimated value of the peak traffic.
[0053]
Next, the operation in the above configuration will be described.
[0054]
FIG. 6 is a flowchart of the operation according to the second embodiment of the present invention.
[0055]
Hereinafter, it is assumed that t = mn. At this time, the time series data is
[0056]
[Equation 14]

Can be expressed as
[0057]
First, {z1,..., Zmn} are divided into m groups by the measurement data dividing unit 210, and the i-th group is divided into {z (i-1) n + 1, z (i-1) n + 2,. （(1 ≦ i ≦ m) (step 201).
[0058]
Next, the maximum value data acquisition unit 220 acquires the maximum value data xi as follows (1 ≦ i ≦ m) (step 202).
[0059]
xi = max {z (i-1) n + 1, z (i-1) n + 2,.
By the way, based on the extreme value theory, the distribution of the maximum value converges to a general extreme value distribution under an appropriate assumption. The general extreme value distribution is a distribution function
[Equation 15]

Is the probability distribution given as follows: Here, ξ, μ, and σ are parameters.
[0061]
(Equation 16)

In the present embodiment, it is assumed that the assumption of the extreme value theory holds, and the maximum value data {x1,..., Xm} follows a general extreme value distribution.
[0062]
Here, the worst value data distribution estimating unit 230 that estimates the probability distribution of the maximum value will be described. First, the average e, the variance v, and the skewness s of the maximum value data are obtained as follows (step 203).
[0063]
[Equation 17]

Then, the parameters of the general extreme value distribution are estimated as 推定, σ, and μ as follows (step 204).
[0064]
Step 204-1) Solve the following equation, and estimate the value of ξ
(Equation 18)

To determine. However,
[0066]
[Equation 19]

It is. Γ (•) is a gamma function.
[0067]
(Equation 20)

Step 204-2) Using the result of step 204-1, an estimated value of σ j is calculated as follows:
(Equation 21)

To determine.
[0069]
(Equation 22)

Step 204-3) Using the results of steps 204-1 and 204-2, an estimated value of μ is obtained as follows:
(Equation 23)

To determine.
[0071]
[Equation 24]

It was found
(Equation 25)

By using the distribution function of the probability distribution according to the maximum value data,
[0073]
(Equation 26)

Is estimated (step 205).
[0074]
Finally, the peak traffic estimation unit 240 will be described. if,
[0075]
[Equation 27]

Is negative, the right end of the probability distribution is bounded,
[0076]
[Equation 28]

Is required. Therefore, this becomes the estimated value of the peak traffic (step 206). on the other hand,
[0077]
(Equation 29)

Is positive, the right end of the probability distribution is not bounded, and the peak traffic is estimated using another method such as a conventional technique (step 207).
[0078]
It is also possible to construct the operations of FIGS. 4 and 6 in the above embodiment as a program and install them on a computer used as a peak traffic estimation device, or distribute them via a network.
[0079]
In addition, the constructed program is stored in a hard disk device connected to a computer used as a peak traffic estimating device, or in a portable storage medium such as a flexible disk or a CD-ROM, and is installed when implementing the present invention. It is also possible.
[0080]
It should be noted that the present invention is not limited to the above embodiment, and various changes and applications are possible within the scope of the claims.
[0081]
【The invention's effect】
As described above, according to the present invention, when measurement data relating to a network of a communication service of interest is obtained, the communication quality index greatly deviates from a normal value without identifying a probability distribution followed by the entire measurement data. By estimating only the value or the stochastic behavior in the case of taking the maximum value, it is possible to efficiently determine an estimated peak traffic value that is more reliable than in the past.
[Brief description of the drawings]
FIG. 1 is a diagram for explaining the principle of the present invention.
FIG. 2 is a principle configuration diagram of the present invention.
FIG. 3 is a configuration diagram of a peak traffic estimation device according to the first embodiment of the present invention.
FIG. 4 is a flowchart of an operation according to the first embodiment of the present invention.
FIG. 5 is a configuration diagram of a peak traffic estimation device according to a second embodiment of the present invention.
FIG. 6 is a flowchart of an operation according to the second embodiment of the present invention.
[Explanation of symbols]
REFERENCE SIGNS LIST 100 Peak traffic estimation device 110 Reference value determination unit 120 Probability distribution estimation unit, probability distribution estimation unit 130 Peak traffic estimation unit, peak traffic estimation unit 140 Input unit 150 Output unit 200 Peak traffic estimation unit 210 Measurement data division unit 220 Maximum value data Acquisition unit 230 Worst value data distribution estimation unit 240 Peak traffic estimation unit 250 Input unit 260 Output unit

Claims (8)

In the peak traffic estimation method for estimating the peak traffic that serves as a guideline when designing, managing and controlling the network of the communication service of interest,
Among the measurement data relating to the network of the communication service of interest input by the input means, the probability distribution according to the data value of which data value exceeds a predetermined reference value is estimated,
A peak traffic estimation method, wherein a right end of the probability distribution is output as an estimated value of peak traffic. In the peak traffic estimation method for estimating the peak traffic that serves as a guideline when designing, managing and controlling the network of the communication service of interest,
The measurement data related to the communication service network of interest input by the input unit is divided into a plurality of groups,
From each of the divided groups, obtain the data that takes the maximum value in the group,
Estimate the probability distribution that the acquired maximum value data follows,
A peak traffic estimation method, wherein a right end of the probability distribution is output as an estimated value of peak traffic. A peak traffic estimating device for estimating peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest,
Probability distribution estimating means for estimating a probability distribution followed by data whose data value exceeds a predetermined reference value, among measurement data relating to the network of the communication service of interest inputted by the input means,
A peak traffic estimating device, comprising: peak traffic estimating means for outputting a right end of the probability distribution estimated by the probability distribution estimating means as an estimated value of peak traffic. A peak traffic estimating device for estimating peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest,
Measurement data division means for dividing the measurement data relating to the network of the communication service of interest inputted by the input means into a plurality of groups,
From each group divided by the measurement data dividing means, maximum value data obtaining means for obtaining data having the maximum value in the group,
A probability distribution estimating means for estimating a probability distribution according to the acquired maximum value data;
Peak traffic estimating means for outputting a right end of the probability distribution as an estimated value of peak traffic. A peak traffic estimation program for estimating peak traffic serving as a guideline when designing, managing and controlling a network of a communication service of interest,
Inputting measurement data relating to the network of the communication service of interest;
Among the measurement data, a probability distribution estimation step of estimating a probability distribution followed by data whose data value exceeds a predetermined reference value,
A peak traffic estimation step in which the right end of the probability distribution estimated by the probability distribution estimation step is an estimated value of the peak traffic,
Outputting the estimated value of the peak traffic. A peak traffic estimation program for estimating peak traffic serving as a guideline when designing, managing and controlling a network of a communication service of interest,
Inputting measurement data relating to the network of the communication service of interest;
A step of dividing the measurement data into a plurality of groups;
From each group divided by the measurement data dividing means, a maximum value data obtaining step of obtaining data having a maximum value in the group,
A probability distribution estimating step of estimating a probability distribution according to the obtained maximum value data;
A peak traffic estimation step of setting the right end of the probability distribution as an estimated value of the peak traffic,
Outputting the estimated value of the peak traffic. A storage medium storing a peak traffic estimation program for estimating a peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data relating to the network of the communication service of interest;
Among the measurement data, a probability distribution estimation step of estimating a probability distribution followed by data whose data value exceeds a predetermined reference value,
A peak traffic estimation step in which the right end of the probability distribution estimated by the probability distribution estimation step is an estimated value of the peak traffic,
Outputting a peak traffic estimation value; and a storage medium storing a peak traffic estimation program. A storage medium storing a peak traffic estimation program for estimating a peak traffic serving as a guideline when designing, managing, and controlling a network of a communication service of interest,
Inputting measurement data relating to the network of the communication service of interest;
A step of dividing the measurement data into a plurality of groups;
From each group divided by the measurement data dividing means, a maximum value data obtaining step of obtaining data having a maximum value in the group,
A probability distribution estimating step of estimating a probability distribution according to the obtained maximum value data;
A peak traffic estimation step of setting the right end of the probability distribution as an estimated value of the peak traffic,
Outputting a peak traffic estimation value; and a storage medium storing a peak traffic estimation program.

Images