Robust Convolutional Neural Networks in SRAM-based FPGAs: a Case Study in Image Classification

Autor: Nilberto H. Medina, Fabio Benevenuti, Adria Barros de Oliveira, Vitor Ângelo Paulino de Aguiar, Fernanda Lima Kastensmidt, M. A. Guazzelli, Nemitala Added

Publikováno v: Journal of Integrated Circuits and Systems. 16:1-12

This work discusses the main aspects of vulnerability and degradation of accuracy of an image classification engine implemented into SRAM-based FPGAs under faults. The image classification engine is an all-convolutional neural-network (CNN) trained w

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::9da7b5af506d778d72be364fd8d757d4
https://doi.org/10.29292/jics.v16i2.504

Investigating the reliability impacts of neutron-induced soft errors in aerial image classification CNNs implemented in a softcore SRAM-based FPGA GPU

Autor: Fabio Benevenuti, Marcio M. Gonçalves, Evaldo Carlos F. Pereira, Rafael G. Vaz, Odair L. Gonçalez, Rodrigo Possamai Bastos, Manon Letiche, Fernanda L. Kastensmidt, José Rodrigo Azambuja

Publikováno v: Microelectronics Reliability. 138:114738

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::99d33c1248476491d8d5a042e7f7cde7
https://doi.org/10.1016/j.microrel.2022.114738

Reliability Calculation With Respect to Functional Failures Induced by Radiation in TMR Arm Cortex-M0 Soft-Core Embedded Into SRAM-Based FPGA

Autor: Adria Barros de Oliveira, Nilberto H. Medina, Fabio Benevenuti, Vitor A. P. Aguiar, M. A. Guazzelli, Fernanda Lima Kastensmidt, Nemitala Added, Luis A. C. Benites

Publikováno v: IEEE Transactions on Nuclear Science. 66:1433-1440

This paper presents comparative results from fault injection (FI) and heavy ions accelerated irradiation on a Xilinx 7 series static RAM (SRAM)-based field-programmable gate array (FPGA) for a soft-core microprocessor mitigated by triple modular redu

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::422a6c09c1c9b829b3fc3dd264eb4afb
https://doi.org/10.1109/tns.2019.2921796

Evaluating softcore GPU in SRAM-based FPGA under radiation-induced effects

Autor: Marcio M. Goncalves, Jose Rodrigo Azambuja, Fernanda Lima Kastensmidt, Michael Hübner, Marcelo Brandalero, Giani Braga, Hector Gerardo Munoz Hernandez, Fabio Benevenuti

Publikováno v: Microelectronics Reliability. 126:114348

This work investigates selective mitigation techniques to improve the reliability of a configurable open-source softcore GPU implemented in an SRAM-based FPGA against configuration memory faults. It explores the unhardened and hardened reliability cu

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::85c16859e4c639bdf1ebb91be7a0365b
https://doi.org/10.1016/j.microrel.2021.114348

Exploiting approximate computing for low-cost fault tolerant architectures

Autor: Alberto Bosio, Juan Fonseca, Fernanda Lima Kastensmidt, Gennaro S. Rodrigues, Fabio Benevenuti

Publikováno v: Proceedings of the 32nd Symposium on Integrated Circuits and Systems Design
Proceedings of the 32nd Symposium on Integrated Circuits and Systems Design, Aug 2019, São Paulo, Brazil. pp.1-6, ⟨10.1145/3338852.3339875⟩
SBCCI

This work investigates how the approximate computing paradigm can be exploited to provide low-cost fault tolerant architectures. In particular, we focus on the implementation of Approximate Triple Modular Redundancy (ATMR) designs using the precision

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1bdd32157e50f14e7a69d1a14724873d
https://hal.archives-ouvertes.fr/hal-02462201