User Activity Recognition for Energy Saving inSmart Homes

Risultato della ricerca: Other

10 Citazioni (Scopus)

Abstract

Current energy demand for appliances in smart homes is nowadays becoming a severe challenge, due to economic and environmental reasons; effective automated approaches must take into account basic information about users, such as the prediction of their course of actions. The present proposal consists in recognizing user daily life activities by simply relying on the analysis of environmental sensory data in order to minimize energy consumption by guaranteeing that peak demands do not exceed a given threshold. Our approach is based on information theory in order to convert raw data into high-level events, used to represent recursively structured activities. Experiments based on publicly available datasets and consumption models are provided to show the effectiveness of our proposal.
Lingua originaleEnglish
Pagine1-9
Numero di pagine9
Stato di pubblicazionePublished - 2013

Fingerprint

Information theory
Energy conservation
Energy utilization
Economics
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications

Cita questo

@conference{dfb0da7c00a7405c8c034912f3f41052,
title = "User Activity Recognition for Energy Saving inSmart Homes",
abstract = "Current energy demand for appliances in smart homes is nowadays becoming a severe challenge, due to economic and environmental reasons; effective automated approaches must take into account basic information about users, such as the prediction of their course of actions. The present proposal consists in recognizing user daily life activities by simply relying on the analysis of environmental sensory data in order to minimize energy consumption by guaranteeing that peak demands do not exceed a given threshold. Our approach is based on information theory in order to convert raw data into high-level events, used to represent recursively structured activities. Experiments based on publicly available datasets and consumption models are provided to show the effectiveness of our proposal.",
keywords = "Energy saving, Pattern Recognition, User Profiling",
author = "Pietro Cottone and Salvatore Gaglio and {Lo Re}, Giuseppe and Marco Ortolani",
year = "2013",
language = "English",
pages = "1--9",

}

TY - CONF

T1 - User Activity Recognition for Energy Saving inSmart Homes

AU - Cottone, Pietro

AU - Gaglio, Salvatore

AU - Lo Re, Giuseppe

AU - Ortolani, Marco

PY - 2013

Y1 - 2013

N2 - Current energy demand for appliances in smart homes is nowadays becoming a severe challenge, due to economic and environmental reasons; effective automated approaches must take into account basic information about users, such as the prediction of their course of actions. The present proposal consists in recognizing user daily life activities by simply relying on the analysis of environmental sensory data in order to minimize energy consumption by guaranteeing that peak demands do not exceed a given threshold. Our approach is based on information theory in order to convert raw data into high-level events, used to represent recursively structured activities. Experiments based on publicly available datasets and consumption models are provided to show the effectiveness of our proposal.

AB - Current energy demand for appliances in smart homes is nowadays becoming a severe challenge, due to economic and environmental reasons; effective automated approaches must take into account basic information about users, such as the prediction of their course of actions. The present proposal consists in recognizing user daily life activities by simply relying on the analysis of environmental sensory data in order to minimize energy consumption by guaranteeing that peak demands do not exceed a given threshold. Our approach is based on information theory in order to convert raw data into high-level events, used to represent recursively structured activities. Experiments based on publicly available datasets and consumption models are provided to show the effectiveness of our proposal.

KW - Energy saving

KW - Pattern Recognition

KW - User Profiling

UR - http://hdl.handle.net/10447/92467

M3 - Other

SP - 1

EP - 9

ER -