Machine Learning-based Intrusion Detection in IoT Systems

IoT device security has become a major concern as these resource-constrained devices are increasingly deployed at scale, limiting the feasibility of advanced onboard security mechanisms. This challenge is particularly critical for Mobile Network Operators (MNOs), who must protect their infrastructure and customers from faulty or malicious IoT devices. To address this, MNOs rely on IoT device fingerprinting to profile deployed devices and support effective intrusion detection, despite having limited control over customer-owned devices. Hence, this project aims to develop robust intrusion detection solutions based on IoT device fingerprinting using machine learning approaches, enabling accurate identification of faulty and malicious behavior at the network level.

Faculty Supervisor:

Habib Louafi

Student:

Partner:

École Polytechnique de Sousse

Discipline:

Computer science

Sector:

Education

University:

Université TÉLUQ

Program: