Reliability of FINN-Generated CNN Accelerators for Image Classification on SRAM-Based FPGAs Under Heavy-Ion-Induced Faults
This study examines the performance of two Convolutional Neural Networks (CNNs) designed for aerial image classification in the presence of radiation-induced bit-flips. We modify these CNNs by adjusting parameters such as quantization and parallelism to facilitate their implementation using the FINN inference engine, which is optimized for the AMD/Xilinx FPGAs. The aim is to evaluate the impact of different quantization levels, network topologies, and architectural parallelism on area, computational performance, and reliability in the presence of soft-errors. Emulated fault injection and heavy ion irradiation were performed. The results indicate that the same CNN topology can exhibit up to a 2.7× difference in Mean Fluence to Failure (MΦTF) by altering quantization and architectural parallelism. The findings demonstrate that higher dependability can be obtained by carefully combining a suitable CNN topology with optimized quantization and architectural parallelism.
Authors
- Eduardo Luiz Augusto Macchione (ORCID: https://orcid.org/0000-0002-9496-9300)
- N. H. Medina (ORCID: https://orcid.org/0000-0003-0650-6507)
- N. Added (ORCID: https://orcid.org/0000-0001-9611-8721)
- Fernanda Lima Kastensmidt (ORCID: https://orcid.org/0000-0001-5767-8582)
- Matheus S. Pereira (ORCID: https://orcid.org/0000-0002-4141-7166)
- Vitor A. P. Aguiar (ORCID: https://orcid.org/0000-0001-6199-0800)
- José Rodrigo Azambuja (ORCID: https://orcid.org/0000-0002-2627-5075)
- Fábio Benevenuti (ORCID: https://orcid.org/0000-0002-0996-9470)
- Greiciane J. Cesário
- Antonio Carlos Schneider Beck (ORCID: https://orcid.org/0000-0002-4492-1747)
- Arthur F. Ely
- Saulo G. Alberton (ORCID: https://orcid.org/0000-0002-7390-3660)
- M. Guazzelli (ORCID: https://orcid.org/0000-0001-7110-7241)
Institutions
- Universidade Federal do Rio Grande do Sul (BR)
- Centro Universitário FEI (BR)
- Universidade Brasil (BR)
Publication Details
- Journal
- IEEE Transactions on Nuclear Science
- Published
- 2025-07-14
- DOI
- https://doi.org/10.1109/tns.2025.3588691
- Primary Topic
- Integrated Circuits and Semiconductor Failure Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Fundação de Amparo à Pesquisa do Estado de São Paulo
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
- Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul
- Office of Naval Research Global