A fuzzy-arithmetic-based reliability assessment model for digital circuits (FARAM-DC)

Reliability is crucial in digital circuits and SRAM-based FPGAs, the programmable devices used for circuit implementation. Hence, this article proposes an assessment method for failure probability and reliability of digital circuits in SRAM-based FPGAs. There are various studies on the reliability of FPGA circuits; most of them consider the effect of cosmic radiation on the circuit's reliability, which results in Single-Event Upsets (SEUs). The impact of hardware and software failures is rarely considered. The proposed method in this research considers the effects of different factors on the circuit's failure, including hardware failures resulting from dust, wetness, temperature, and jitter, software failures resulting from programming errors, and SEUs resulting from radiation. Since precise estimation of the factors’ effects on the circuit's failure is difficult, qualitative values are used to explain the impacts and estimate the failure probability and reliability using Fuzzy arithmetic. The results present a precise evaluation of the reliability of digital circuits, which, ignoring each effective factor, leads to an overestimation of reliability.