Partially Observable Multivariate Point Processes with Linear Random Compensators: Analysis and Filtering with Applications to Queueing Networksстатья
Информация о цитировании статьи получена из
Scopus
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 28 февраля 2020 г.
Автор:
Borisov A.V.
Журнал:
IFAC-PapersOnLine
Том:
48
Номер:
11
Год издания:
2015
Издательство:
Elsevier Ltd
Местоположение издательства:
London
Первая страница:
1108
Последняя страница:
1113
DOI:
10.1016/j.ifacol.2015.09.342
Аннотация:
The paper is devoted to the optimal filtering problem of the Markov jump state given the multivariate point observations. The investigated observation system is set in terms of the martingale representation. The distinctive feature of the observations is in their compensator, which is a linear random transformation of the unobservable state trajectory. Basing on the predictable martingale characteristics the mutual distribution of the state and observations is reconstructed. The obtained filtering estimate is determined via the finite closed system of the ordinary linear differential equations with a random right-hand side and recursive algebraic relations. The presented theoretical results are applied to the modelling of the service processes in some standard partially observable queueing networks. The paper also contains comparison of the presented filtering algorithm with the classical cases of Kalman-Bucy and Wonham. В© 2015
Добавил в систему:
Борисов Андрей Владимирович