Approche déterministe de l'acquisition comprimée et la reconstruction des signaux issus de capteurs intelligents distribués

Abstract : A wireless body area network (WBAN) is a new class of wireless networks dedicated to monitor human physiological parameters. It consists of small electronic devices, also called nodes, attached to or implanted in the human body. Each node comprises one or many sensors which measure physiological signals, such as electrocardiogram or body heat, and the characteristics of the surrounding environment. These nodes are mainly subject to a significant energy constraint due to the fact that the miniaturization has reduced the size of their batteries. A solution to minimize the energy consumption would be to compress the sensed data before wirelessly transmitting them. Indeed, research has shown that most of the available energy are consumed by the wireless transmitter. Conventional compression methods are not suitable for WBANs because they involve a high computational power and increase the energy consumption. To overcome these limitations, we use compressed sensing (CS) to compress and recover the sensed data. We propose a simple and efficient encoder to compress the data. We also introduce a new algorithm to reduce the complexity of the recovery process. A partnership with TEA (Technologie Ergonomie Appliquées) company allowed us to experimentally evaluate the performance of the proposed method during which a numeric version of the encoder has been used. We also developed and validated an analog version of the encoder.
Document type :
Theses
Complete list of metadatas

https://hal.univ-lorraine.fr/tel-01751944
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 1:29:16 PM
Last modification on : Tuesday, August 21, 2018 - 9:46:04 AM

File

DDOC_T_2015_0136_RAVELOMANANTS...
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01751944, version 1

Citation

Andrianiaina Ravelomanantsoa. Approche déterministe de l'acquisition comprimée et la reconstruction des signaux issus de capteurs intelligents distribués. Autre. Université de Lorraine, 2015. Français. ⟨NNT : 2015LORR0136⟩. ⟨tel-01751944⟩

Share

Metrics

Record views

199

Files downloads

90