AUTHOR=Xie Yu , Qiao Tingting , Xie Yizhuang , Chen He 

TITLE=Soft error mitigation and recovery of SRAM-based FPGAs using brain-inspired hybrid-grained scrubbing mechanism

JOURNAL=Frontiers in Computational Neuroscience

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/computational-neuroscience/articles/10.3389/fncom.2023.1268374

DOI=10.3389/fncom.2023.1268374

ISSN=1662-5188

ABSTRACT=Soft error has increasingly become a critical concern for SRAM-based Field Programmable Gate Arrays (FPGAs), which could corrupt the configuration memory that store configuration data describing the custom-designed circuit architecture. To mitigate this kind of errors, this paper proposes a brain-inspired hybrid-grained scrubbing mechanism consisting fine-grained and coarse-grained scrubbing to mitigate and repair the errors as quickly as possible after a SEU occurrence. Inspired by the human brain's ability to filter out redundant and irrelevant information, we propose a mechanism that can mask invalid position information when errors occur. Comparing with scrubbing of full configuration memory, this mechanism can achieve precise error location and recovery utilizing targeted scrubbing of specific frames or modules. The effectiveness is evaluated by executing fault injection campaigns on International Symposium on Circuits and Systems 1989 (ISCAS89) benchmark circuits and Fault Tolerant Fast Fourier transform (FT-FFT) circuit. If upsets are detected, it will be repaired with fine-grained or coarse-grained scrubbing depending on its location. The experiment results show that this mechanism can effectively mitigate and repair single bit upset (SBU) and double bit upsets (DBU). In addition, the mechanism is proven to be superior in error recovery time and hardware overhead compared to counterpart approaches.SRAM-based FPGAs have been widely used in security and mission-critical applications, such as Wang et al. (2015), González et al. (2011), Hartley et al. (2013) and Yang & Fathy (2009). Due to their high logic density, low power consumption, reconfiguration feature and parallel computing capability, especially in aerospace and avionic domains. However, they are extremely sensitive to radiation effects.