| KEYWORD |
GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA
KataOS for CVA6 microprocessor in IoT devices
keywords DISTRIBUTED SYSTEM, INTERNET OF THINGS, IOT
Reference persons LUCA BARBIERATO, EDOARDO PATTI
Research Groups DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, EDA Group, ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ICT4SS - ICT FOR SMART SOCIETIES
Thesis type EXPERIMENTAL
Description KataOS is a provably secure operating system for embedded ML hardware developed by Google's AmbiML team. It is based on the seL4 microkernel and implemented almost entirely in Rust.
KataOS is designed to address the challenges of managing the privacy and security of data collected by smart devices. It provides a verifiably secure platform that protects the user's privacy because it is logically impossible for applications to breach the kernel's hardware security protections, and the system components are verifiably secure.
KataOS is also highly efficient and lightweight, making it ideal for use on embedded devices with limited resources. It is also relatively easy to develop, thanks to its use of Rust.
Here are some examples of potential use cases for KataOS:
- Smart home devices: KataOS could be used to develop secure smart home devices that can collect and process data without compromising the user's privacy.
- Wearable devices: KataOS could be used to develop secure wearable devices that can collect and process health and fitness data without compromising the user's privacy.
- Industrial IoT devices: KataOS could be used to develop secure industrial IoT devices that can collect and process data from sensors and actuators without compromising the safety and security of the underlying industrial systems.
The thesis primarily aims to examine the architecture of the KataOS operating system. Subsequently, it seeks to formulate a strategy for adapting KataOS to a System-on-Chip (SoC) featuring a 64-bit RISC-V core. The suggestion is to utilise an SoC created at the University of Bologna, which incorporates a CVA6 core.
Deadline 03/11/2025
PROPONI LA TUA CANDIDATURA