Andrea Portaluri
Study and Development of an Isolation-based Approach for Mitigating Radiation-induced Errors in Reconfigurable Systems-on-Chip.
Rel. Luca Sterpone, Sarah Azimi, Corrado De Sio. Politecnico di Torino, Corso di laurea magistrale in Mechatronic Engineering (Ingegneria Meccatronica), 2021
|
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview |
Abstract: |
In the last years, Reconfigurable Systems-on-Chip and SRAM-based Field Programmable Gate Arrays have been widely adopted for mission-critical tasks in aerospace, avionics and automotive, mainly thanks to their flexibility and low costs. Although, one of their main drawbacks is the high susceptibility to radiations-charged particles, both in space and at sea level. Among the several mitigation techniques, Isolation Design Flow obtains promising results in terms of reliability improvement and required time to implement it by acting only at the floorplanning phase. However, when considering numerous blocks and complex systems, implementation of state-of-the-art Isolation Design Flow can lead to a difficult (or even impossible) placement stage. The main objective pursed by this thesis is the analysis and development of domains-based Isolation Design Flow guidelines to ease the floorplanning phase in the case of hardening-by-replication systems, where the application of the state-of-the-art approach would be otherwise very complex and time-consuming. Thus, a fault injection platform has been built in order to perform experimental campaigns of validation on such a technique. These design rules have been, then, applied to a Triple Modular Redundant benchmark implemented on a Xilinx Zynq-7000 AP-SoC to quantify its effectiveness by means of fault injection campaigns. In addition to this, fault injection campaigns have also been carried out on the state-of-the-art Isolation Design Flow to compare the results. |
---|---|
Relatori: | Luca Sterpone, Sarah Azimi, Corrado De Sio |
Anno accademico: | 2021/22 |
Tipo di pubblicazione: | Elettronica |
Numero di pagine: | 65 |
Soggetti: | |
Corso di laurea: | Corso di laurea magistrale in Mechatronic Engineering (Ingegneria Meccatronica) |
Classe di laurea: | Nuovo ordinamento > Laurea magistrale > LM-25 - INGEGNERIA DELL'AUTOMAZIONE |
Aziende collaboratrici: | Politecnico di Torino |
URI: | http://webthesis.biblio.polito.it/id/eprint/20518 |
Modifica (riservato agli operatori) |