6533b7d7fe1ef96bd1268fff
RESEARCH PRODUCT
Experimental trade-offs between different strategies for multihop communications evaluated over real deployments of wireless sensor network for environmental monitoring
Juan J. Pérez-solanoAntonio Soriano-asensiSantiago Felici-castellJaume Segura-garciaMiguel Garcia-pinedasubject
020203 distributed computingInternetComputer Networks and Communicationsbusiness.industryComputer scienceTrade offsGeneral Engineering020206 networking & telecommunications02 engineering and technologyXarxes locals sense fil Wi-Filcsh:QA75.5-76.95Environmental monitoring0202 electrical engineering electronic engineering information engineeringlcsh:Electronic computers. Computer scienceInternet of ThingsbusinessWireless sensor networkComputer networkdescription
Although much work has been done since wireless sensor networks appeared, there is not a great deal of information available on real deployments that incorporate basic features associated with these networks, in particular multihop routing and long lifetimes features. In this article, an environmental monitoring application (Internet of Things oriented) is described, where temperature and relative humidity samples are taken by each mote at a rate of 2 samples/min and sent to a sink using multihop routing. Our goal is to analyse the different strategies to gather the information from the different motes in this context. The trade-offs between ‘sending always’ and ‘buffering locally’ approaches were analysed and validated experimentally, taking into account power consumption, lifetime, efficiency and reliability. When buffering locally, different options were considered such as saving in either local RAM or FLASH memory, as well different alternatives to reduce overhead with different packet sizes. The conclusion is that in terms of energy and durability, the best option is to reduce the overhead. Nevertheless, sending larger packets is not worthy when the probability of retransmission is high. If real-time monitoring is required, then sending always is better than buffering locally.
year | journal | country | edition | language |
---|---|---|---|---|
2018-05-01 |