Towards in-network data prediction in wireless sensor networks

  • Authors:
  • Tales Benigno Matos;Angelo Brayner;Jose Everardo Bessa Maia

  • Affiliations:
  • University of Fortaleza;University of Fortaleza;State University of Ceara

  • Venue:
  • Proceedings of the 2010 ACM Symposium on Applied Computing
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Researches in Wireless Sensor Networks (WSN) have been focused on saving power in sensor nodes. An efficient strategy to achieve this goal is to reduce the amount of data sent through the network, In this work, we propose an efficient strategy that predicts data in WSNs aiming at reducing the data traffic in WSNs and thus maximizing the network lifetime. The proposed prediction strategy, denoted ADAGA-P, is based on a linear regression model, using data acquired from one or several sensors. ADAGA-P is executed in an in-network fashion by several sensors geographically distributed in a WSN. Furthermore, in ADAGA-P the regression model is adjusted dynamically, if any sensor in the WSN senses an "outlier". Experimental results are presented using real data. These results show that the proposed strategy reduces energy consumption in WSNs.