PhD - Leuven | More than two weeks ago
Develop in depth understanding of a mechanism behind fault attacks induced state change in MRAM and FeRAM memory arrays for an application in cryptography.
Voltage gate assisted spin orbit torque magnetic random-access memory (VGSOT-MRAM) is one of these emerging non-volatile memories, that has a potential for use in embedded cache, microcontroller units (MCU), and embedded flash applications. On the other hand, also ferroelectric random-access memory (Fe-RAM) has a potential to replace the existing dynamic random-access memories.
Side channel, micro-architectural and fault attacks are a major challenge in cached and embedded flash security. A famous example is the Rowhammer attack on DRAMs1. In row hammering attacks repeated access to the rows results in flipping of bits in adjacent rows, which allows untrusted code to escape its sandbox. As the density of memory cells increases, and new types of memories appear, also new types of Rowhammering attacks are becoming a persistent threat to chip security.
In this Ph.D., the aim is to study the memory cells and array switching behavior beyond normal operating conditions. The student will perform in-depth fault attacks and investigate whether state changes can be induced on MRAM and FeRAM cell and arrays. The student will try to model the experimental results to understand the fault induced switching mechanisms. Eventually, the student can investigate row hammering and clock/voltage fault attacks on memory arrays controlled by FPGA boards. From the physcial analysis the student can deduce the countermeasures to improve the memory security.
1 Google Online Security Blog: Introducing Half-Double: New hammering technique for DRAM Rowhammer bug (googleblog.com)
Required background: M.Sc. Physics, M.Tech. Electronics/Electrical Engineering, M.Tech. Computer Science, M.Tech. Cyber Security.
Type of work: 20% literature study, 40% experimental work, 40% Simulation/modelling.
Supervisor: Ingrid Verbauwhede
Daily advisor: Ankit Kumar, Robin Degraeve
The reference code for this position is 2023-010. Mention this reference code on your application form.