Many examples of applications require real-time computing, such as nuclear power plants, automotive, railway, and avionics systems, air traffic control, robotics, and military systems. In such scenarios, high predictability and security must be guaranteed in all operating conditions since all monitoring, control, and actuation functions are implemented in software and executed under stringent timing constraints. Responding to the persistent demand for more processing power and motivated by the mentioned reasons, hardware vendors have commercialized new heterogeneous multi-core systems-on-a-chip. Therefore, the design of next-generation safety-critical systems necessitates software advancements that, on the other hand, give rise to several unique challenges that have only recently started to be addressed in the research community. Motivated by the strong field knowledge provided by Rete Ferroviaria Italiana S.p.A. (the Italian railway infrastructure manager), who partially funded the research, the thesis focuses on mixed-critical systems, especially those designed for transportation. The study aims to investigate and develop novel software mechanisms and architectures for the next generation of safety-critical systems. Most of the well-known challenges rising when executing trusted and untrusted applications on the same platform have been addressed by taking advantage of novel technological features, such as virtualization capabilities offered by the underlying hardware. Through the use of state-of-the-art development techniques, assisted by hardware accelerators and virtualization, the security of a system can be significantly increased, and the efficiency can be improved by sharing the hardware resources available while reducing the costs of implementing the system but, at the same, meeting stringent system certification requirements.
Mechanisms for safe and secure mixed-criticality real-time embedded systems
SERRA, GABRIELE
2024
Abstract
Many examples of applications require real-time computing, such as nuclear power plants, automotive, railway, and avionics systems, air traffic control, robotics, and military systems. In such scenarios, high predictability and security must be guaranteed in all operating conditions since all monitoring, control, and actuation functions are implemented in software and executed under stringent timing constraints. Responding to the persistent demand for more processing power and motivated by the mentioned reasons, hardware vendors have commercialized new heterogeneous multi-core systems-on-a-chip. Therefore, the design of next-generation safety-critical systems necessitates software advancements that, on the other hand, give rise to several unique challenges that have only recently started to be addressed in the research community. Motivated by the strong field knowledge provided by Rete Ferroviaria Italiana S.p.A. (the Italian railway infrastructure manager), who partially funded the research, the thesis focuses on mixed-critical systems, especially those designed for transportation. The study aims to investigate and develop novel software mechanisms and architectures for the next generation of safety-critical systems. Most of the well-known challenges rising when executing trusted and untrusted applications on the same platform have been addressed by taking advantage of novel technological features, such as virtualization capabilities offered by the underlying hardware. Through the use of state-of-the-art development techniques, assisted by hardware accelerators and virtualization, the security of a system can be significantly increased, and the efficiency can be improved by sharing the hardware resources available while reducing the costs of implementing the system but, at the same, meeting stringent system certification requirements.File | Dimensione | Formato | |
---|---|---|---|
PhD_Thesis_Serra.pdf
embargo fino al 01/06/2026
Dimensione
3.4 MB
Formato
Adobe PDF
|
3.4 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/217480
URN:NBN:IT:SSSUP-217480