Interview with Nikolaos Georgantas, MiMove team leader
We met with Nikolaos Georgantas, leader of the new MiMove team since 1 July 2014.
What does MiMove mean and what are its principal areas of research?
MiMove stands for Middleware on the Move: middleware for new generation mobile distributed systems is the core of our research. These systems pose new challenges that call for radical new methods in the design, development and operation of systems.
First, globalisation and the omnipresence of IT infrastructures and networks (Internet and the cloud) and personal mobile devices (smartphones and tablets) create operating environments which are extremely rich in terms of functionalities while highly dynamic and unpredictable for systems. Such a diversity can never be entirely anticipated when systems are designed. This is why we study mobile distributed systems that “emerge” according to their environment and which evolve along with it while they're operating.
These complex environments integrating millions of sensors and actuators can be embedded in any connected physical object (Internet of Things). Personal mobile devices with numerous sensors in particular constitute a large-scale physical sensing infrastructure with limitless possibilities. We are interested in mobile sensing and actuation systems composed of very large ephemeral populations of collaborative entities.
In addition to physical sensing, the widespread embrace of personal devices and their use at any moment of daily life gives individuals the opportunity to participate actively in sensing the real world around them. In particular, this concerns phenomena that are impossible to detect only through embedded sensors (e.g., congestion at a location may be unpleasant in a tube station and the opposite at a concert). We are studying social mobile crowd-sensing systems, i.e., systems that can be used to collect and use big data from the population.
Is this basic or applied research? Are there concrete applications?
Our research effort covers theoretical and applied aspects of mobile distributed systems. We adopt a “systems” approach, which includes all phases from system design to implementation of prototypes and experimental assessment. Such an approach also implies a collaborative effort with researchers from different fields.
To be more precise, we design new models, algorithms and protocols to demonstrate and manage system characteristics. We then analyse and validate them theoretically to demonstrate their accuracy. We also implement the proposed functionalities, including the design of middleware architecture and development of corresponding middleware instances while taking into account actual system constraints. Finally, we assess the systems experimentally to ensure that they are representative of reality.
The results of our research can be applied to applications in a variety of fields. The smart city is our prime target, since it is a field of growing technological, economic and social interest. Within the smart city, the entire urban space is connected, monitored and managed; the objective is the sustainable development of the city and the well-being of its citizens. This includes data management, and may bring about radically new ways of life through the social involvement of citizens.
In this context, our research on mobile distributed systems, particularly physical and social sensing, may contribute significantly to collecting and sharing data about the city and the commitment of citizens to the life of the city. Since this field is highly multidisciplinary, we seek collaborations with researchers from other IT specialisations as well as other disciplines, such as the social sciences.
Do you have partners in industry and academia?
We have both academic and industrial partners, particularly for collaborative projects at the national, European and international levels.
MiMove has initiated an Inria International Project Lab – currently being evaluated – called CityLab, dedicated to smart cities. CityLab includes eight other Inria project teams: Clime, Dice, Fun, Myriads, Oak, Smis, Urbanet and Willow. CityLab also works in close collaboration with universities in California, including UC Berkeley, as part of the Inria@SiliconValley program.
We are also involved with the EIT ICT Labs 3Cixty project, which covers intelligent transportation systems, and includes partners such as DFKI, UCL, Politechnico di Milano, Eurecom, Thales, Telecom Italia and Ambientic.
At the national level, we are part of the ANR Murphy Project on sensor reliability in conjunction with CNAM and LAAS.
Lastly, we are also starting to collaborate with USC on smart cities, and we have created two teams associated with UFG Brazil and IIIT-Delhi.