Semantic Interoperability for IoT Platforms in Support of Decision Making: An Experiment on Early Wildfire Detection.

Nikos Kalatzis,Yiorgos Marinellis,George Routis,Marios Avgeris,Symeon Papavassiliou,Ioanna Roussaki,Miltiades Anagnostou

doi:10.3390/s19030528

Nikos Kalatzis, Yiorgos Marinellis + Show 5 more

Open Access

https://doi.org/10.3390/s19030528

Copy DOI

Abstract

One of the main obstacles towards the promotion of IoT adoption and innovation is data interoperability. Facilitating cross-domain interoperability is expected to be the core element for the realisation of the next generation of the IoT computing paradigm that is already taking shape under the name of Internet of Everything (IoE). In this article, an analysis of the current status on IoT semantic interoperability is presented that leads to the identification of a set of generic requirements that act as fundamental design principles for the specification of interoperability enabling solutions. In addition, an extension of NGSIv2 data model and API (de-facto) standards is proposed aiming to bridge the gap among IoT and social media and hence to integrate user communities with cyber-physical systems. These specifications have been utilised for the implementation of the IoT2Edge interoperability enabling mechanism which is evaluated within the context of a catastrophic wildfire incident that took place in Greece on July 2018. Weather data, social media activity, video recordings from the fire, sensor measurements and satellite data, linked to the location and the time of this fire incident have been collected, modeled in a uniform manner and fed to an early fire detection decision support system. The findings of the experiment certify that achieving minimum data interoperability with light-weight, plug-n-play mechanisms can be realised with significant benefits for our society.

Highlights

The Internet of Things (IoT) is reaching new levels of popularity and maturity with high penetration in various application domains directly influencing the societal and economical aspects of our lives
As this major incident drew the attention of media, it was feasible to retrospectively collect valuable data and utilise them for evaluating the interoperability and decision making approaches introduced in this article
Social media posts, video streams, and weather data were collected and different IoT2Edge-Agents where employed in order to process and translate this data according to the NGSIv2 data model

Summary

Introduction

The Internet of Things (IoT) is reaching new levels of popularity and maturity with high penetration in various application domains directly influencing the societal and economical aspects of our lives. IoT growth is explosive and there are already billions of connected smart objects, embedded systems, microcontrollers and sensors that have penetrated our world connecting home users, businesses, public facilities and enterprise systems with one another. One of the critical issues is that today’s IoT landscape consists of platforms and proprietary systems that are mainly isolated and act as “vertical silos”. These silos impede the creation of cross-domain, cross-platform

Methods

Results

Conclusion