Personal webpage
Today’s System-on-Chip (SoC) include advanced features that allow for dynamic energy-management to better adapt to current computation loads, heat, and required performance and energy constraints. However, an emergent category of attacks has shown that these features can be maliciously exploited to perform internal attacks (i.e., attacks performed by software co-located with sensitive applications on the same hardware), even circumventing Trusted Execution Environments (TEE) security guarantees. Reported attacks include covert channels, Side-Channel Attacks (SCA) and Fault Injection (FI) by exploiting the frequency, voltage, temperature, current, and power supply. Although totally disabling energy-management mechanisms would limit and avoid most vulnerabilities, this would go against the performance and energy optimizations, achieved gains, and might lead to significant overheads. Instead of limiting or disabling these optimization mechanisms, in CoPhyTEE, we propose to benefit from them to built a security governor able to detect at run-time possible vulnerable execution scenarios,
This is within the collaboration with ANSSI. Scientific Advisory Board includes KU Leuven and Radboud University.
Scientific conference article accepted at CASCADE: A Lightweight Embedded Detection System against Voltage Drop Fault Attacks in Multi-Tenant FPGAs, Le Gonidec, G., Bouffard, G., Prevotet, J. C., & Méndez Real, M. (2026). Springer Constructive Approaches for SeCurity Analysis and Design of Embedded systems (CASCADE).
Scientific journal article accepted at ACM TECS: Do Not Trust Power Management: A Survey on Internal Energy-based Attacks Circumventing Trusted Execution Environments Security Properties, Le Gonidec, G., Bouffard, G., Prevotet, J. C., & Méndez Real, M. (2025). ACM Transactions on Embedded Computing Systems (TECS), 24(4), 1-35. open-version available here.
Presentation and poster at JAIF (Journée thématique sur les Attaques par Injection de Fautes), available here; presentation, poster.
This project is fund by the Franch Agency of Research ANR. It runs from 2023 - 2027.