Affiliated project IoT-SDC
Affiliated CEA MITIC/Université Gaston Berger and SMIA/Université d'Abomey-Calvi
Title: Dynamic framework for preventing and responding to security incidents in the IoT
Start date : 01/06/2021
The effects of the Internet of Things (IoT) are not going unnoticed, as they have turned our lives upside down, making it easier for people to work in all fields, including military, environmental, agricultural, medical, industrial and domestic. Indeed, connecting objects not only enables us to acquire as much information as we want, but also to act remotely (turning a light on or off, closing doors, parking cars, consulting a patient, etc.).
However, the multiplication of mobile applications and services, and advances in telecommunications networks, are having a major impact on the IoT, due to the spectacular increase in data flows across networks. The question then arises as to how to manage all this data, and how to classify it so as to retain the most essential.
Furthermore, in the field of cyber security, the emergence of the IoT exposes many objects and devices to the risk of becoming, if they are not already, a thingbot - a botnet integrating independent connected objects - or zombie connected objects. This is where an intelligent classifier comes into its own: by analyzing data, it can create a dynamic framework for finding correlations between threats and events, so as to be able to react immediately if an incident and/or suspected attack is detected.
Objectives / Expected results :
Our objective in this project is to find a data classification model in the IoT for better data management.
As a solution, we propose to insert neural networks as data classifiers in the IoT.
In this way, we are pursuing a dual objective in a multi-disciplinary research project targeting applications in cyber-security on the one hand, and food safety on the other.
Finally, these objectives are not limitative, as other fields of application are also possible. These include home automation, agriculture, the military, meteorology, etc.
Contribution / added value to the affiliated project :
The security aspect is a major challenge in IoT infrastructures. Our contribution will consist in setting up a Framework capable of preventing security incidents in the IoT, but also of responding to them when necessary. Our approach to achieving this goal is to use artificial intelligence and deep learning techniques.
Thesis supervisor Prof. Chérif DIALLO, CEA MITIC, Senegal
Thesis co-supervisor Prof. Eugene EZIN, CEA SMIA, Benin
Other contributors to thesis supervision : Prof. Mady CISSE, ESP, UCAD, Senegal