Energy eficiente synchronization for alarm driven wireless sensor networks

Many applications of wireless sensor networks require that nodes, besides monitoring a given phenomenon, must be able to detect and communicate asynchronous events (e.g. alarms), implying that they have to often listen to the medium in idle mode, which is inherently energy wasteful. In such a scenar...

Descripción completa

Detalles Bibliográficos
Autor: Nadas, João Pedro Battistella
Tipo de recurso: tesis de maestría
Estado:Versión publicada
Fecha de publicación:2016
País:Brasil
Institución:Universidade Tecnológica Federal do Paraná (UTFPR)
Repositorio:Repositório Institucional da UTFPR (da Universidade Tecnológica Federal do Paraná (RIUT))
Idioma:inglés
OAI Identifier:oai:repositorio.utfpr.edu.br:1/2309
Acceso en línea:http://repositorio.utfpr.edu.br/jspui/handle/1/2309
Access Level:acceso abierto
Palabra clave:Energia - Consumo
Detectores
Sistemas de comunicação sem fio
Métodos de simulação
Engenharia elétrica
Energy consumption
Detectors
Wireless communication systems
CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::TELECOMUNICACOES
Engenharia Elétrica
Descripción
Sumario:Many applications of wireless sensor networks require that nodes, besides monitoring a given phenomenon, must be able to detect and communicate asynchronous events (e.g. alarms), implying that they have to often listen to the medium in idle mode, which is inherently energy wasteful. In such a scenario time synchronization is crucial to efficiently operate in duty-cycles and minimize energy consumption. In this work we assess the impact of the trade-off between spending energy with more frequent synchronizations and in return saving it by reducing the idle listening window necessary for the desired reliability of the communication. The optimal frequency of time synchronizations is obtained analytically and corroborated by numerical results, showing that several times less overall energy may be spent with a finer synchronization when compared with maintaining the minimum clock precision required by the phenomenon being monitored, greatly extending the life-span of the network.Furthermore, a closed form upper bound to this optimal number is derived by approximating transmit power being of much more significance when compared to receive power. Using this result, we predict and then simulate that this optimal number will be increased by the listening power, the number of times which a node has to listen to the medium idly, the level of confidence at which the system is designed to work, the synchronization interval and the variance of the relative oscillation frequency between synchronizing nodes. On the other hand, this number will be smaller when the energy cost of synchronization is higher (e.g. when active communication energy increases).