Scientists create unified platform to manage different types of IoT sensors

^{Translation by Diana Taborda}

Initially, the Internet of Things (IoT) relied mainly on real (physical) sensors, but today it uses data from other types of sensors, namely virtual and social, making management extremely complex. To overcome this problem, a team from the Faculty of Science and Technology, University of Coimbra (FCTUC), has developed a platform that allows the unified management of heterogeneous environments supporting the IoT.

Today, rather than the Internet of Things, we have the Internet of Everything, because it's not just the Internet of Things, but also the Internet of People and the Internet of Software. There are a series of sensors from different, heterogeneous sources that collect huge amounts of information that allow us to measure, among other parameters, temperature, humidity and air pollution, but also human movements, industrial environments, especially in Industry 4.0, etc.; for example, our mobile phones are devices full of sensors that, without us even noticing, measure all and everything," explains Fernando Boavida, co-author of the study published in the scientific journal IEEE Internet of Things.

Thus, the current IoT paradigm, which makes it possible to connect the Internet to different types of sensors and environments in order to collect data on a large scale, requires information management systems, i.e. systems that allow data to be controlled and monitored, regardless of the type of sensor.

There are several types of sensors: real sensors, virtual (software) sensors and social sensors, also known as human sensors, because they extract information posted by people on social networks. According to the professor and researcher at FCTUC's Department of Computer Engineering, the starting question for developing the platform was "How can we efficiently and effectively manage different types of IoT data sources?

Taking advantage of several studies on IoT that he has been conducting for more than a decade, together with Jorge Sá Silva of FCTUC's Department of Electrical and Computer Engineering, the team set out to find solutions that could answer this question.

"Based on technologies widely used in IoT to manage real sensors, specifically FIWARE technology, we developed a computing solution capable of managing different types of sensors and analysing the best management protocols in terms of performance," says Fernando Boavida, explaining that in practice "we took protocols and technologies used to manage real sensors and adapted them to virtual and social sensors".

The prototype of the system developed was then tested for a semester with volunteers, students from the University of Coimbra and a polytechnic school in Ecuador, demonstrating the viability of the proposed approach. The results obtained represent a further step in the development of the Internet of Things: "With the same tools, it is possible to have a unified management of different types of sensors, that is, heterogeneous environments of the Internet of Things. It is a platform that allows the effective management of data, regardless of the type of sensor - physical, virtual and social," says Fernando Boavida.

If you think about a smart city, for example, "where you can infer the quality of life in a city, you have to manage a lot of factors: environmental factors like temperature, pollution, humidity, etc.; more complex factors like how easy it is to move around the city; and you have to understand if people are satisfied. In this example, different types of sensors are involved. A system like ours makes it possible to manage all this in a unified way," says the FCTUC professor.

The next phase of research will focus on the development of social sensors, which represent the greatest challenge because the information that can be extracted is virtually unlimited. In particular, Fernando Boavida and Jorge Sá Silva's team will study the processing and analysis of data extracted from social networks. In other words, concludes Fernando Boavida, "we want to close the loop. As well as extracting information from social networks, we also want to provide useful information, advice and data that can help improve people's quality of life".