Consiglio Nazionale delle Ricerche

Tipo di prodottoArticolo in rivista
TitoloHuman activity recognition using multisensor data fusion based on Reservoir Computing
Anno di pubblicazione2016
Formato
  • Elettronico
  • Cartaceo
Autore/iPalumbo F.; Gallicchio C.; Pucci R.; Micheli A.
Affiliazioni autoriCNR-ISTI, Pisa, Italy; Department of Computer Science, University of Pisa, Pisa, Italy; Department of Computer Science, University of Pisa, Pisa, Italy; Department of Computer Science, University of Pisa, Pisa, Italy
Autori CNR e affiliazioni
  • FILIPPO PALUMBO
Lingua/e
  • inglese
AbstractActivity recognition plays a key role in providing activity assistance and care for users in smart homes. In this work, we present an activity recognition system that classifies in the near real-time a set of common daily activities exploiting both the data sampled by sensors embedded in a smartphone carried out by the user and the reciprocal Received Signal Strength (RSS) values coming from worn wireless sensor devices and from sensors deployed in the environment. In order to achieve an effective and responsive classification, a decision tree based on multisensor data-stream is applied fusing data coming from embedded sensors on the smartphone and environmental sensors before processing the RSS stream. To this end, we model the RSS stream, obtained from a Wireless Sensor Network (WSN), using Recurrent Neural Networks (RNNs) implemented as efficient Echo State Networks (ESNs), within the Reservoir Computing (RC) paradigm. We targeted the system for the EvAAL scenario, an international competition that aims at establishing benchmarks and evaluation metrics for comparing Ambient Assisted Living (AAL) solutions. In this paper, the performance of the proposed activity recognition system is assessed on a purposely collected real-world dataset, taking also into account a competitive neural network approach for performance comparison. Our results show that, with an appropriate configuration of the information fusion chain, the proposed system reaches a very good accuracy with a low deployment cost.
Lingua abstractinglese
Altro abstract-
Lingua altro abstract-
Pagine da87
Pagine a107
Pagine totali21
RivistaJournal of ambient intelligence and smart environments (Print)
Attiva dal 2009
Editore: IOS Press - Amsterdam
Paese di pubblicazione: Paesi Bassi
Lingua: inglese
ISSN: 1876-1364
Titolo chiave: Journal of ambient intelligence and smart environments (Print)
Titolo proprio: Journal of ambient intelligence and smart environments. (Print)
Numero volume della rivista8
Fascicolo della rivista2
DOI10.3233/AIS-160372
Verificato da refereeSì: Internazionale
Stato della pubblicazionePublished version
Indicizzazione (in banche dati controllate)
  • ISI Web of Science (WOS) (Codice:000373206800002)
  • Scopus (Codice:2-s2.0-84962580238)
Parole chiaveAAL, Activity recognition, Neural networks, Reservoir Computing, Sensor data fusion, WSN
Link (URL, URI)http://content.iospress.com/articles/journal-of-ambient-intelligence-and-smart-environments/ais372
Titolo parallelo-
Licenza-
Scadenza embargo-
Data di accettazione-
Note/Altre informazioni-
Strutture CNR
  • ISTI — Istituto di scienza e tecnologie dell'informazione "Alessandro Faedo"
Moduli/Attività/Sottoprogetti CNR
  • ICT.P07.008.002 : Tecnologie e sistemi wireless eterogenei interconnessi
Progetti Europei-
Allegati
Human activity recognition using multisensor data fusion based on Reservoir Computing (documento privato )
Descrizione: published version - Codice PuMa: cnr.isti/2016-A0-022
Tipo documento: application/pdf
Human activity recognition using multisensor data fusion based on Reservoir Computing
Descrizione: postprint version
Tipo documento: application/pdf