Architecture and algorithm for data fusion of heterogeneous multi source information in internet of things (IoT)‎ paradigm

Abstract EN

Ubiquitous computing that is enabled by the Internet of Things (IoT) paradigm effectuates several aspects of human life, both technologically and socially.

These 'Things' have to use highly accurate, reliable, and precise information to be smarter.

This can be accomplished by efficient IoT architecture that is coupled with a data fusion (DF) algorithm.

DF can be defined as the technique of fusing the information coming from multiple sources which are representing the same real-world objects into precise and intelligent information.

In this work, we propose a multi-hop IoT architecture which is helpful in facilitating DF and is also energyefficient.

This architecture achieves this by using a multi-hop mechanism.

Furthermore, we develop a DF algorithm, Two Phase Data Fusion (2PDF) algorithm.

We simulated a system for the Air Quality measurement application, and then we compared the results of the 2PDF algorithm with the Kalman filter (KF) algorithm, which is a benchmark algorithm.

Results showed that the 2PDF outperformed the KF in the case of outlier data.

This work can be considered as the first step in the field of IoT architecture dependent DF algorithms.

Furthermore, the proposed work helps to see DF in a broader context of IoT.