Saltar a casilla de búsquedaSaltar a navegaciónIr directamente al contenido principal

A deep learning approach to device-free people counting from WiFi signals

  • Iker Sobron(autor correspondiente)
    ,
  • ,
  • Iñaki Eizmendi
    ,
  • Manuel Velez
*Autor correspondiente de este trabajo
  • Basque Center for Applied Mathematics
Publicación:
Capítulo del libro/informe/acta de congreso
Capítulo
Investigación
Revisión por expertos

Abstract

<p>The last decade has witnessed a progressive interest shown by the community on inferring the presence of people from changes in the signals exchanged by deployed wireless devices. This non-invasive approach finds its rationale in manifold applications where the provision of counting devices to the people expected to traverse the scenario at hand is not affordable nor viable in the practical sense, such as intrusion detection in critical infrastructures. A trend in the literature has focused on modeling this paradigm as a supervised learning problem: a dataset with WiFi traces and their associated number of people is assumed to be available a priori, which permits to learn the pattern between traces and the number of people by a supervised learning algorithm. This paper advances over the state of the art by proposing a novel convolutional neural network that infers such a pattern over space (frequency) and time by rearranging the received I/Q information as a three-dimensional tensor. The proposed layered architecture incorporates further processing elements for a better generalization capability of the overall model. Results are obtained over real WiFi traces and compared to those recently reported over the same dataset for shallow learning models. The superior performance shown by the model proposed in this work paves the way towards exploring the applicability of the latest advances in Deep Learning to this specific case study.</p>

Research Output Information

Tipo de resultado

Publicación:
Capítulo del libro/informe/acta de congreso
Capítulo
Investigación
Revisión por expertos

Idioma original

Inglés

Páginas desde-hasta (Número de páginas)

Páginas 275-286 (12 páginas)

Hitos de publicación

  • Publicada - 2018

Estado de publicación

Publicada - 2018

Editorial

Springer Verlag, Alemania

Serie de publicación

  • Nombre de serie de publicación: Studies in Computational Intelligence
    ISSN (Impreso): 1860-949X
    Volumen: 798

ID de publicación

  • Scopus: 85053463659

Título de publicación principal

Studies in Computational Intelligence

Funding Details

Acknowledgements. This work was supported in part by the Spanish Ministry of Economy and Competitiveness under project 5GnewBROS (TEC2015-66153-P MINECO/FEDER, EU) and by the Basque Government (IT683-13 and the EMAITEK program). This work was supported in part by the Spanish Ministry of Economy and Competitiveness under project 5GnewBROS (TEC2015-66153-PMINECO/FEDER, EU) and by the Basque Government (IT683-13 and the EMAITEK program).
FundersFunding numbers
Spanish Ministry of Economy and Competitiveness
TEC2015-66153-P MINECO/FEDER
EC
-
FEDER
-
Eusko Jaurlaritza
IT683-13
MINECO
-
EU
-

Publication metrics

Métricas

SciVal
citaciones
6
SciVal
FWCI
1.11
SciVal
Número de autores
4
SciVal
Percentil de artículo
74

PlumX, se abre en una pestaña nueva

Capturas
6
Número de citas
6