Title:	Leakage Estimation for Physical Unclonable Functions
DNr:	LiU-gpu-2025-5
Project Type:	LiU Compute
Principal Investigator:	Onur Günlü <onur.gunlu@liu.se>
Affiliation:	Linköpings universitet
Duration:	2025-05-01 – 2026-01-01
Classification:	20299
Keywords:

Abstract

This is a master's project that has no direct commercial purpose. We will use polar codes and neural leakage estimators to generate secret keys from digital circuits. Our project uses Deep Neural Networks to estimate the entropy and mutual information of multivariate binary random variables. ----- NEW PART ----- The Master's student is now a research assistant and we are continuing our research on deep neural networks, so we would like to have an extension until 2026-01-01.