Enhancing dependability through profiling in the collaborative internet of things

The future of the Internet of Things (IoT) is the Collaborative Internet of Things (C-IoT) in which different IoT deployments collaborate to provide better services. For instance, in smart city scenarios, C-IoT will have the potential to provide immersive multimedia user-experiences based on content and context fusion, immersive multi-sensory environments, location-based and media internet technologies, and augmented reality. However, this future paradigm will only be possible if the right decisions can be made based on the analysis of huge volumes of collected data: i.e. if the dependability of C-IoT is ensured. To address this challenge, we studied a simplified view of a C-IoT architecture composed of devices using three different technologies that have enabled the existence of IoT (RFID, NFC and Beacons). However, our proposal could be extended to any other devices in the context of C-IoT. To enhance the dependability of C-IoT, we deploy statistical data analysis techniques to improve the quality of the data obtained from identification and sensing devices and to select the most reliable devices that provide trusted (i.e. non-faulty) data in order to support accurate decision-making.