Sign CUSUM Algorithm for Change-Point Detection of the MMPP Controlling Chain State; Communications in Computer and Information Science; Vol. 638 : Information Technologies and Mathematical Modelling - Queueing Theory and Applications, ITMM 2016

Dettagli Bibliografici
Parent link:Communications in Computer and Information Science
Vol. 638 : Information Technologies and Mathematical Modelling - Queueing Theory and Applications, ITMM 2016.— 2014.— [P. 18-33]
Autore principale: Burkatovskaya Yu. B. Yuliya Borisovna
Ente Autore: Томский политехнический университет Институт кибернетики, ИК
Altri autori: Kbanova T. V. Tatjyana Vladimirovna, Tokareva O. S. Olga Sergeevna
Riassunto:Title screen
The authors consider the Markov modulated Poisson process with two states of the Markovian controlling chain. The flow intensity of the observed process depends on the unobserved controlling chain state. All the process parameters are supposed to be unknown. The paper develops a new sequential change-point detection method based on the cumulative sum control chart approach to determine the switching points of the flow intensity. Usage of special sign statistics allows the obtaining of theoretical characteristics of the proposed algorithm.
Режим доступа: по договору с организацией-держателем ресурса
Lingua:inglese
Pubblicazione: 2014
Soggetti:
Accesso online:http://dx.doi.org/10.1007/978-3-319-44615-8_2
Natura: Elettronico Capitolo di libro
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=650785
Descrizione
Riassunto:Title screen
The authors consider the Markov modulated Poisson process with two states of the Markovian controlling chain. The flow intensity of the observed process depends on the unobserved controlling chain state. All the process parameters are supposed to be unknown. The paper develops a new sequential change-point detection method based on the cumulative sum control chart approach to determine the switching points of the flow intensity. Usage of special sign statistics allows the obtaining of theoretical characteristics of the proposed algorithm.
Режим доступа: по договору с организацией-держателем ресурса
DOI:10.1007/978-3-319-44615-8_2