FAROES: Fairness And Reliability using Overlay Expenseless Set-out for duty-cycle optimization in WSN

International audience Wireless sensor networks (WSNs) consist of a large number of entities that collaborate in order to provide given services. Unfortunately, due to their tiny size, these entities cannot be equipped with a long-life battery. In order to minimize overall energy consumption and max...

Full description

Bibliographic Details
Main Authors: Bonomi, Silvia, Busnel, Yann, Baldoni, Roberto, Prakash, Ravi
Other Authors: Middleware Laboratory (MIDLAB), Università degli Studi di Roma "La Sapienza" = Sapienza University Rome, Department of Computer Science Dallas (University of Texas at Dallas), University of Texas at Dallas Richardson (UT Dallas)
Format: Conference Object
Language:English
Published: HAL CCSD 2009
Subjects:
Online Access:https://hal.archives-ouvertes.fr/hal-00480995
https://hal.archives-ouvertes.fr/hal-00480995/document
https://hal.archives-ouvertes.fr/hal-00480995/file/pdccs09.pdf
Description
Summary:International audience Wireless sensor networks (WSNs) consist of a large number of entities that collaborate in order to provide given services. Unfortunately, due to their tiny size, these entities cannot be equipped with a long-life battery. In order to minimize overall energy consumption and maximize the lifetime of the network, a widely accepted approach is to implement a \emph{duty cycle}. Specifically, in order to save energy, sensors are not active all the time, and switch off their capabilities according to a specific schedule. In this paper, we propose a generic method to schedule these duty-cycles, taking into account two important challenges in WSNs: \emph{strong reliability} of the network and \emph{fairness} between sensors. To achieve these aims, we ensure reliability by deploying several $k$-connected overlays, which became active one after the other. The fairness rule is ensured by spreading the sensors into overlay according to specific characteristics as energy consumption, network density, \emph{etc.} We then propose different heuristics and probability models to achieve our outcome, and finally validate them through numerical evaluation