PUMA
Istituto di Scienza e Tecnologie dell'Informazione     
Urfalioglu O., Kuruoglu E. E., Cetin E. A. Ardısık Monte Carlo Yontemiyle Bindirilmis Olay Sezimi = Superimposed Event Detection by Sequential Monte Carlo Methods. In: IEEE 15th Signal Processing and Communication Applications Conference (Eskisehir, Turchia, 11-13 June 2007). Proceedings, vol. 1 pp. 1268 - 1271. IEEE Turkey Signal Processing Chapter (ed.). Anadolu University Press, 2007.
 
 
Abstract
(English)
In this paper1, we consider the detection of rare events by applying particle filtering. We model the rare event as an AR signal superposed on a background signal. The activation and deactivation times of the AR-signal are unknown. We solve the online detection problem of this superpositional rare event by extending the state space dimension by one. The additional parameter of the state represents the AR-signal, which is zero when deactivated. Numerical experiments demonstrate the effectiveness of our approach.
Abstract
(Italiano)
In turco: Bu bildiride, seyrek olayların (rare events) sezimi problemi icin parc¸acık s¨uzgec¸lerine dayanan bir y¨ontem onermekteyiz.C¸alıs¸mamızda seyrek olaylar, arkaplan is¸aretine bindirilmis¸ bir ¨ozba˘glanımlı s¨urec¸ (AR) olarak modellenmis¸lerdir. O¨zbag˘lanımlı su¨recin etkinles¸tirme ve etkinsizles¸tirme zamanları bilinmemektedir. Bindirilmis¸ seyrek olayın gerc¸ek zamanda sezim problemi durum uzayı boyutunu genis¸leterek c¸ ¨oz¨ulm¨us¸t¨ur. Ek durum parametresi etkinsizles¸tirme durumunda sıfır olan AR-is¸aretini temsil etmektedir. Sayısal deneyler yaklas¸ımımızın bas¸arımını g¨ostermektedir.
URL: http://siu2007.anadolu.edu.tr/en/index.php
DOI: 10.1109/SIU.2007.4298796
Subject Rare event detection
Sequential Monte Carlo
Particle filtering
G.3 Probability and Statistics. Probabilistic algorithms (including Monte Carlo)
G.3 Probability and Statistics. Time series analysis
60G35 Applications (signal detection, filtering, etc.)
60G70 Extreme value theory; extremal processes
65C35 Stochastic particle methods


Icona documento 1) Download Document PDF


Icona documento Open access Icona documento Restricted Icona documento Private

 


Per ulteriori informazioni, contattare: Librarian http://puma.isti.cnr.it

Valid HTML 4.0 Transitional