SRAM-based field programmable gate arrays (FPGAs) are vulnerable to a single event upset (SEU), which is induced by radiation effect. Therefore, the dependable design techniques become important, and the accurate dependability analysis method is required to demonstrate their robustness. Most of present analysis techniques are performed by using full reconfiguration to emulate the soft error. However, it takes long time to analyze the dependability because it requires many times of reconfiguration to complete the soft error injection. In the present paper, we construct the soft error estimation system to analyze the reliability and to reduce the estimation time. Moreover, we apply monte carlo simulation to our approach, and identify trade-off between accuracy of error rate and estimation time. As a result of our experimentation for 8-bit full-adder and multiplier, we can show the dependability of the implemented system. Also, the constructed system can reduce the estimation time. According to the result, when performing about 50% circuit monte carlo simulation, the error rate is within 20%.
[1]
Takuji Nishimura,et al.
Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator
,
1998,
TOMC.
[2]
Sergio D'Angelo,et al.
A fault injection tool for SRAM-based FPGAs
,
2003,
9th IEEE On-Line Testing Symposium, 2003. IOLTS 2003..
[3]
Luigi Carro,et al.
Fault-Tolerance Techniques for SRAM-Based FPGAs (Frontiers in Electronic Testing)
,
2006
.
[4]
Massimo Violante,et al.
A new reliability-oriented place and route algorithm for SRAM-based FPGAs
,
2006,
IEEE Transactions on Computers.
[5]
Luigi Carro,et al.
Fault-Tolerance Techniques for SRAM-Based FPGAs
,
2006
.