A computational scheme to predict dynamics in IoT systems by using particle filter. (21st March 2017)

• Record Type: Journal Article
• Authors: Cuomo, Salvatore; De Michele, Pasquale; Pragliola, Monica
• Other Names: Zbakh Mostapha guestEditor.; Bakhouya Mohamed guestEditor.; Essaaidi Mohamed guestEditor.; Piccialli Francesco guestEditor.; Chianese Angelo guestEditor.; Jung Jason J. guestEditor.
• Is Part Of: Concurrency and computation. Volume 29:Number 11(2017)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2017-03-21
• Subjects: clustering -- linear multistep method -- particle filter -- stochastic models; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.4101
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: Extract information from data, coming from the real world, is a very fascinating challenge. In the Internet of Things society, sensors, devices, and tools are able to generate a lot of data that could be used to predict behaviours. In this paper, we propose a computational scheme in which the clustering methodology is used to classify information that are adopted as observations of an evolutionary method. A sampling techniques are used to model the unknown of a dynamical system as a random variable, and available information are interpreted as probability density function. The probability density function is approximated by an ensemble of weighted particles. Finally, by using this methodology, we present results on the forecast and track users' behaviours in 2 real‐world case studios.