Traffic data analysis based on extreme value theory and its applications

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

It is important to predict serious deterioration of the telecommunication quality. The purpose of this paper is to predict such serious events by analyzing only a "short" period (i.e., a "small" amount) of teletraffic data. To achieve this end, this paper presents a method for analyzing the tail distributions of variables concerning teletraffic states, because tail distributions are suitable to represent serious events. This method is based on Extreme Value Theory (EVT), which provides a firm theoretical foundation for the analysis. To be more precise, in this paper, we use throughput data measured on an actual network in daily busy hours for 15 minutes, and use its first 10 seconds (known data) to analyze the tail distribution. Then, we evaluate how well the obtained tail distribution can predict the tail distribution of the remaining 890 seconds (unknown data). The result shows that the obtained tail distribution based on EVT by analyzing the small amount of known data can predict the tail distribution of the unknown data much better than methods based on an empirical distribution and log-normal distribution. Furthermore, we apply the obtained tail distribution to predict the peak throughput in unknown data. The results of this paper enable us to predict serious events with lower measurement cost.

Original languageEnglish
Title of host publicationGLOBECOM'04 - IEEE Global Telecommunications Conference
Pages1418-1424
Number of pages7
Volume3
Publication statusPublished - 2004
Externally publishedYes
EventGLOBECOM'04 - IEEE Global Telecommunications Conference - Dallas, TX, United States
Duration: 2004 Nov 292004 Dec 3

Other

OtherGLOBECOM'04 - IEEE Global Telecommunications Conference
CountryUnited States
CityDallas, TX
Period04/11/2904/12/3

Fingerprint

Throughput
Normal distribution
Telecommunication
Deterioration
Costs

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Uchida, M. (2004). Traffic data analysis based on extreme value theory and its applications. In GLOBECOM'04 - IEEE Global Telecommunications Conference (Vol. 3, pp. 1418-1424)

Traffic data analysis based on extreme value theory and its applications. / Uchida, Masato.

GLOBECOM'04 - IEEE Global Telecommunications Conference. Vol. 3 2004. p. 1418-1424.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Uchida, M 2004, Traffic data analysis based on extreme value theory and its applications. in GLOBECOM'04 - IEEE Global Telecommunications Conference. vol. 3, pp. 1418-1424, GLOBECOM'04 - IEEE Global Telecommunications Conference, Dallas, TX, United States, 04/11/29.
Uchida M. Traffic data analysis based on extreme value theory and its applications. In GLOBECOM'04 - IEEE Global Telecommunications Conference. Vol. 3. 2004. p. 1418-1424
Uchida, Masato. / Traffic data analysis based on extreme value theory and its applications. GLOBECOM'04 - IEEE Global Telecommunications Conference. Vol. 3 2004. pp. 1418-1424
@inproceedings{758a90ff569941e88d4865bd3dae7a03,
title = "Traffic data analysis based on extreme value theory and its applications",
abstract = "It is important to predict serious deterioration of the telecommunication quality. The purpose of this paper is to predict such serious events by analyzing only a {"}short{"} period (i.e., a {"}small{"} amount) of teletraffic data. To achieve this end, this paper presents a method for analyzing the tail distributions of variables concerning teletraffic states, because tail distributions are suitable to represent serious events. This method is based on Extreme Value Theory (EVT), which provides a firm theoretical foundation for the analysis. To be more precise, in this paper, we use throughput data measured on an actual network in daily busy hours for 15 minutes, and use its first 10 seconds (known data) to analyze the tail distribution. Then, we evaluate how well the obtained tail distribution can predict the tail distribution of the remaining 890 seconds (unknown data). The result shows that the obtained tail distribution based on EVT by analyzing the small amount of known data can predict the tail distribution of the unknown data much better than methods based on an empirical distribution and log-normal distribution. Furthermore, we apply the obtained tail distribution to predict the peak throughput in unknown data. The results of this paper enable us to predict serious events with lower measurement cost.",
author = "Masato Uchida",
year = "2004",
language = "English",
volume = "3",
pages = "1418--1424",
booktitle = "GLOBECOM'04 - IEEE Global Telecommunications Conference",

}

TY - GEN

T1 - Traffic data analysis based on extreme value theory and its applications

AU - Uchida, Masato

PY - 2004

Y1 - 2004

N2 - It is important to predict serious deterioration of the telecommunication quality. The purpose of this paper is to predict such serious events by analyzing only a "short" period (i.e., a "small" amount) of teletraffic data. To achieve this end, this paper presents a method for analyzing the tail distributions of variables concerning teletraffic states, because tail distributions are suitable to represent serious events. This method is based on Extreme Value Theory (EVT), which provides a firm theoretical foundation for the analysis. To be more precise, in this paper, we use throughput data measured on an actual network in daily busy hours for 15 minutes, and use its first 10 seconds (known data) to analyze the tail distribution. Then, we evaluate how well the obtained tail distribution can predict the tail distribution of the remaining 890 seconds (unknown data). The result shows that the obtained tail distribution based on EVT by analyzing the small amount of known data can predict the tail distribution of the unknown data much better than methods based on an empirical distribution and log-normal distribution. Furthermore, we apply the obtained tail distribution to predict the peak throughput in unknown data. The results of this paper enable us to predict serious events with lower measurement cost.

AB - It is important to predict serious deterioration of the telecommunication quality. The purpose of this paper is to predict such serious events by analyzing only a "short" period (i.e., a "small" amount) of teletraffic data. To achieve this end, this paper presents a method for analyzing the tail distributions of variables concerning teletraffic states, because tail distributions are suitable to represent serious events. This method is based on Extreme Value Theory (EVT), which provides a firm theoretical foundation for the analysis. To be more precise, in this paper, we use throughput data measured on an actual network in daily busy hours for 15 minutes, and use its first 10 seconds (known data) to analyze the tail distribution. Then, we evaluate how well the obtained tail distribution can predict the tail distribution of the remaining 890 seconds (unknown data). The result shows that the obtained tail distribution based on EVT by analyzing the small amount of known data can predict the tail distribution of the unknown data much better than methods based on an empirical distribution and log-normal distribution. Furthermore, we apply the obtained tail distribution to predict the peak throughput in unknown data. The results of this paper enable us to predict serious events with lower measurement cost.

UR - http://www.scopus.com/inward/record.url?scp=18144425461&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18144425461&partnerID=8YFLogxK

M3 - Conference contribution

VL - 3

SP - 1418

EP - 1424

BT - GLOBECOM'04 - IEEE Global Telecommunications Conference

ER -