Soft-Error Vulnerability Estimation Approach Based on the SET Susceptibility of Each Gate

Soft-Error Vulnerability (SEV) is a parameter used to evaluate the robustness of a circuit to the induced Soft Errors (SEs). There are many techniques for SEV estimation, including analytical, electrical and logic simulations, and emulation-based approaches. Each of them has advantages and disadvant...

Full description

Saved in:
Bibliographic Details
Published in:Electronics (Basel) 2019-07, Vol.8 (7), p.749
Main Authors: Batagin Armelin, Fábio, Naviner, Lírida Alves de Barros, d’Amore, Roberto
Format: Article
Language:English
Subjects:
Circuits
Computer Science
Electronics
Engineering Sciences
Hardware Architecture
Ionizing radiation
Logic
Methods
Probability
Radiation
Radiation effects
Simulation
Soft errors
Topology
Transistors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Soft-Error Vulnerability (SEV) is a parameter used to evaluate the robustness of a circuit to the induced Soft Errors (SEs). There are many techniques for SEV estimation, including analytical, electrical and logic simulations, and emulation-based approaches. Each of them has advantages and disadvantages regarding estimation time, resources consumption, accuracy, and restrictions over the analysed circuit. Concerning the ionising radiation effects, some analytical and electrical simulation approaches take into account how the circuit topology and the applied input patterns affect their susceptibilities to Single Event Transient (SET) at the gate level. On the other hand, logic simulation and emulation techniques usually ignore these SET susceptibilities. In this context, we propose a logic simulation-based probability-aware approach for SEV estimation that takes into account the specific SET susceptibility of each circuit gate. For a given operational scenario, we extract the input patterns applied to each gate and calculate its specific SET susceptibility. For the 38 analysed benchmark circuits, we obtained a reduction from 15.27% to 0.68% in the average SEV estimation error, when comparing the estimated value to a reference obtained at the transistor level. The results point out an improvement of the SEV estimation process by considering the specific SET susceptibilities.
ISSN:2079-9292
2079-9292
DOI:10.3390/electronics8070749