A Data Fusion Algorithm for Multiple Applications in Wireless Sensor Networks

Abstract

Wireless sensor networks (WSN) are core components of the Internet of Things paradigm. Traditionally, WSNs are designed for a specific application. However, in the Internet of Things era, the sensing and network infrastructure should be shared by a set of applications from multiple owners. In such a scenario, the massive amount of data produced by the widely spread sensors is processed and analyzed to produce value-added information for the end user. By sharing the same infrastructure with multiple users, the application requirements (for instance, data intervals or the potential events of interest) may not be known a priori. In this chapter, we present a solution to properly integrate data from multiple applications without the knowledge about specific application requirements and uncover useful information from such data. We present Hephaestus, an entropic information fusion algorithm, which uses mean, kurtosis and skewness to apply a heuristic that divides the dataset into multiple features, for multiple applications. In our results, Hephaestus achieved high accuracy while incurring in low overhead for the resource-constrained devices of WSNs.