In the last two decades, data and information fusion has experienced significant development due mainly to advances in sensor technology. The sensors provide a continuous flow of data about the environment in which they are deployed, which is received and processed to build a dynamic estimation of the situation. With current technology, it is relatively simple to deploy a set of sensors in a specific geographic area, in order to have highly sensorized spaces. However, to be able to fusion and process the information coming from the data sources of a highly sensorized space, it is necessary to solve certain problems inherent to this type of technology. The challenge is analogous to what we can find in the field of the Internet of Things (IoT). IoT technology is characterized by providing the infrastructure capacity to capture, store, and process a huge amount of heterogeneous sensor data (in most cases, from different manufacturers), in the same way that it occurs in data fusion applications. This work is not simple, mainly due to the fact that there is no standardization of the technologies involved (especially within the communication protocols used by the connectable sensors). The solutions that we can find today are proprietary solutions that imply an important dependence and a high cost. The aim of this paper is to present a new open source platform with capabilities for the collection, management and analysis of a huge amount of heterogeneous sensor data. In addition, this platform allows the use of hardware-agnostic in a highly scalable and cost-effective manner. This platform is called Thinger.io. One of the main characteristics of Thinger.io is the ability to model sensorized environments through a high level language that allows a simple and easy implementation of data fusion applications, as we will show in this paper.
Keywords: IoT middleware; data fusion applications; scalabillity.