Skip to Main content Skip to Navigation
Conference papers

A Bayesian approach for an efficient data reduction in IoT

Abstract : Nowadays, Internet of Things (IoT) coupled with cloud computing begins to take an important place in economic systems and in society daily life. It has got a large success in several application areas, ranging from smart city applications to smart grids. Despite the apparent success, one major challenge that should be addressed is the huge amount of data generated by the sensing devices. The transmission of these huge amount of data to the network may affect the energy consumption of sensing devices, and can also cause network congestion issues. To face this challenge, we present a Bayesian Inference Approach (BIA), which allows avoiding the transmission of high spatio-temporal correlated data. In this paper, BIA is based on a hierarchical architecture with smart nodes, smart gateways and data centers. Belief Propagation algorithm has been chosen for performing an approximate inference on our model in order to reconstruct the missing sensing data. BIA is evaluated based on the data collected from the M3 sensors deployed in the FIT IoT-LAB platform and according to different scenarios. The results show that our proposed approach reduces drastically the number of transmitted data and the energy consumption, while maintaining an acceptable level of data prediction accuracy.
Document type :
Conference papers
Complete list of metadata
Contributor : Jean Cristanel Razafimandimby Connect in order to contact the contributor
Submitted on : Tuesday, December 5, 2017 - 2:24:48 PM
Last modification on : Monday, December 7, 2020 - 9:50:02 AM


Files produced by the author(s)


  • HAL Id : hal-01620373, version 1



Cristanel Razafimandimby, Valeria Loscrì, Anna Maria Vegni, Driss Aourir, Alessandro Neri. A Bayesian approach for an efficient data reduction in IoT. InterIoT 2017 - 3rd EAI International Conference on Interoperability in IoT, Nov 2017, Valencia, Spain. pp.1-7. ⟨hal-01620373⟩



Record views


Files downloads