DocumentCode :
2452943
Title :
SEU and SET fault injection models for fault tolerant circuits
Author :
Petrovic, Vladimir ; Ilic, Marko ; Schoof, Gunter ; Stamenkovic, Zoran
Author_Institution :
IHP, Frankfurt Oder, Germany
fYear :
2012
fDate :
3-5 Oct. 2012
Firstpage :
73
Lastpage :
76
Abstract :
In this paper we describe the fault injection models for the systems tolerant to the single event upsets (SEU) and single event transients (SET). The presented fault modelling approach is based on the random generated SEU faults in sequential logic and SET faults in combinational logic. The fault injection models for triple modular redundant (TMR) and dual modular redundant (DMR) circuits are developed in order to simulate the fault-tolerant systems. The analytical models for calculation of the probability of failure-free TMR and DMR circuits are defined too. To justify the reduced redundancy concept, the simulated and calculated probabilities of failure-free TMR and DMR circuits are presented and discussed. The obtained results show a better trade-off between hardware overhead and circuit failure-free probability of the DMR concept.
Keywords :
combinational circuits; failure analysis; radiation hardening (electronics); sequential circuits; SET fault injection model; SEU fault injection model; circuit failure-free probability; combinational logic; dual modular redundant circuits; failure-free DMR circuits; failure-free TMR circuits; fault modelling approach; fault tolerant circuits; fault-tolerant systems; hardware overhead; probability; random generated SEU faults; sequential logic; single event transients; single event upsets; triple modular redundant circuits; Circuit faults; Integrated circuit modeling; Libraries; Probability; Shift registers; Single event upset; Tunneling magnetoresistance;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronics Conference (BEC), 2012 13th Biennial Baltic
Conference_Location :
Tallinn
ISSN :
1736-3705
Print_ISBN :
978-1-4673-2775-6
Type :
conf
DOI :
10.1109/BEC.2012.6376818
Filename :
6376818
Link To Document :
بازگشت