Smart Embedded Systems
CELL is a multidisciplinary team working in the field of Electronics, Signal and Image Processing for the Modeling and Design of Reconfigurable, Communicating, Reliable and Intelligent Embedded Systems.
A particular emphasis is put on the design and use of Machine Learning techniques in contexts specific to data analysis and data processing problems, with a focus on the explainability and transparency of such algorithms.
Applications include social media content and graph analytics, semantic web data integration and analysis, pattern detection in image, video and multimodal data, search engines for different types of data, knowledge extraction from data, data privacy, etc.
The application domains are vast, such as cultural heritage, security, opinion mining, health, etc.
- An annotated wing interferential pattern dataset of dipteran insects of medical interest for deep learning
- Species identification of phlebotomine sandflies using deep learning and wing interferential pattern (WIP)
- Deep learning and wing interferential patterns identify Anopheles species and discriminate amongst Gambiae complex species
CELL research is now divided into three scientific axes representing the two levels of activity of the team, namely an upstream technological level and an application level linked to the socio-economic fabric. These are addressing problems with societal stakes in line with the areas of expertise and values of the team.
This axis focuses on “Green Communication” issues, in harness with the activities of the IoT platform (emerging technologies of networks of interconnections on a chip). This axis also tackles issues related to component reliability and the problem of “dirty RF” when improving the efficiency of power structures in communicating systems (5G).
Smart Embedded Systems
This research highlights the dual aspect of “Unconventional Sensors” and “Smart Embedded Processing” developed by the team in the team’s flagship application areas (IoT, Telecom, Health, Autonomous Vehicle). This axis addresses both hardware and data processing issues within the constraints of embedded systems.
The activities of this axis are centred on the robustness of heterogeneous embedded systems to environmental disturbances in particular. More generally it also adresses energy and error correction codes in the context of communicating systems.
Through these axes, the team is developing national and international collaborations that allow it to carry out ambitious projects in the fields of Security of communicating systems, Health, and Autonomous vehicles, financed by the ANR, the region, and local calls for projects (CY Initiative, ENSEA, CNRS).
The team’s partners are both academic, companies, through CIFRE theses for example, and from the ecosystem of local partners.
Smart Embedded Systems for Health
Non Conventional Sensors
The Wizarde Project
Reliability and ECC
Circuits and RF Systems
To meet the constraints related to the emergence of new standards in communications systems (ICT, security, defense, space, …), new RF devices or architectures are study and design in this research activity.
Imaging database of daily movement activities
As part of three theses, we develop an artificial intelligence-based fall risk monitoring prototype based on the recording of a person’s daily activities via several sensors
The other ETIS teams
Information, Communication and Imaging
Modelling of Learning
Multimedia Indexing & Data Integration