Conference Paper

Low-power occupancy sensor for an ambient assisted living system

Book Series cp
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
  • Volumen: 9108
  • Fecha: 01 January 2015
  • Páginas: 20-29
  • ISSN: 16113349 03029743
  • ISBN: 9783319188324
  • Source Type: Book Series
  • DOI: 10.1007/978-3-319-18833-1_3
  • Document Type: Conference Paper
  • Publisher: Springer Verlag service@springer.de
© Springer International Publishing Switzerland 2015 In this work, we introduce an Ambient Assisted Living (AAL) system that allows to infer a potential dangerous action of an elderly person living alone at home. This inference is obtained by a specific sensorisation with sensor nodes and a reasoning layer embedded in a personal computer that learns of the users behavior patterns and advices when actual one differs significantly in the normal patterns. In this type of systems, energy is a limited resource therefore sensor devices need to be properly managed to conserve energy. In this paper, a force-capacitive transducer based sensor has been proposed, implemented and tested. This sensor is based on Electro-Mechanical Films (EMFiTM) transducer which is able to detect force variations in a quasi-passive way. The transducer is a capacitor with variable capacity depending on the force exerted on its surface. The characterization of the transducer conducted by us in thisway is not present in the literature. This detection of force is used to trigger an activemechanism to measure the weight by means of the transducer capacity, now modelled by us. A low-power wireless sensor node prototype including this new sensor has been assembled and validated with a wide range of weights. The occupancy detection was successful and the power consumption of the node was increased at less than a 15%, which is acceptable for implementation.

Author keywords

    Indexed keywords

      Funding details