ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-08282007-163800


Tipo di tesi
Tesi di laurea specialistica
Autore
Mancini, Francesca
URN
etd-08282007-163800
Titolo
Gestione del consumo energetico in reti di sensori per il monitoraggio del manto nevoso
Dipartimento
INGEGNERIA
Corso di studi
INGEGNERIA INFORMATICA
Relatori
Relatore Anastasi, Giuseppe
Relatore Alippi, Cesare
Relatore Prof. Marcelloni, Francesco
Parole chiave
  • sampling rate
  • sensing
  • campionamento adattivo
  • risparmio energetico
  • reti di sensori
Data inizio appello
02/10/2007
Consultabilità
Non consultabile
Data di rilascio
02/10/2047
Riassunto
This thesis has been carried out in the PerLab laboratory of the Department of Computer Engineering of Pisa University with the collaboration of Polytechnic Institute of Milan.
It regards the planning and development of an Adaptive Sampling Algorithm for Wireless Sensor Networks.
The algorithm presented in this thesis dinamically estimates the optimal sampling frequency of the signal to be monitored and in turn, it minimizes the activity of both the sensor and the radio, while maintaining an acceptable accuracy on the acquired data.
It employes a methodology quite general that can be applied to sensor node with sensing power consumption equal or larger than that of the radio subsystem. That is the case of multifrequency D-ICE sensor, created by Polytechnic Institute of Turin, in collaboration with Polytechnic Institute of Milan, free from any power-aware optimization technique. The sensor is used to monitor snow composition in the mountain slopes for avalanche forecasting.
To test the Adaptive Sampling Algorithm efficiency and its robustness, we have run preliminary analysis with MATLab software using a simple star topology.
Then we used TOSSIM simulation tool with two-hop cluster topology.
Experiments realized with TOSSIM show that the suggested solution can reduce the number of samples to be acquired and transmitted by 66-79%, with respect to the traditional approach based on fixed over-sampling, while maintening the error at acceptable levels. Reducing the number of samples we gain in term of sensing, computing and trasmission energy.
File