Electrical characterization of memristors for neural networks (AI)

Resistive memories have emerged as a pivotal advancement in modern electronics, offering remarkable advantages in energy efficiency, scalability, and non volatile operation. However, TiO2 based OxRAM devices still face key challenges, particularly regarding circuit level variability and the limited number of demonstrations in realistic application settings. This project addresses these gaps by thoroughly characterizing OxRAM switching dynamics, assessing their integration within analog and digital circuits, and developing automated Python based testing tools on the LOTUS neuromorphic platform. Together, these efforts will help establish the viability of OxRAM based analog modules for neuromorphic processing and next generation embedded AI systems.

Faculty Supervisor:

Malak Kanso

Student:

Partner:

Université Evry Paris-Saclay

Discipline:

Engineering

Sector:

Education

University:

Université de Sherbrooke

Program:

Globalink Research Award

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Find Projects