| KEYWORD |
Ai-Driven Detection of Soft Errors Through Stack and Heap Monitoring
keywords ARTIFICIAL INTELLIGENCE, COMPUTER ARCHITECTURES, PROGRAMMING, RELIABILITY
Reference persons STEFANO DI CARLO, ALESSANDRO SAVINO
External reference persons Enrico Magliano
Research Groups DAUIN - GR-24 - SMILIES - reSilient coMputer archItectures and LIfE Sci
Thesis type RESEARCH / EXPERIMENTAL
Description Goals: Develop an AI-based fault detector targeting real-time operating systems working evaluating heap and stack program occupation.
Description: Radiation-induced soft errors are among the most challenging reliability issues in safety-critical real-time embedded systems (SACRES), such as autonomous driving and aerospace. Typically addressed by various flavors of double modular redundancy (DMR) techniques. This solution is becoming unaffordable due to the complexity of modern microprocessors in all domains. This thesis proposes an AI-powered model to detect soft errors by analyzing stack and heap memory usage, offering a smarter alternative to DMR. It will explore the feasibility of this approach and identify the best AI techniques for accurate fault detection.
Learned Outcomes: real-time operating systems, fault injection, Soft Error Detection, Deep Learning model
Required skills C and Python programming, Operating Systems concepts, basic AI knowledge.
Deadline 12/12/2025
PROPONI LA TUA CANDIDATURA