As LSI chips integrate more transistors and the operating power supply voltage decreases, LSI chips are becoming more vulnerable to the soft error caused by neutrons induced from cosmic rays. The soft error is detected by comparing the duplicated operation results in double modular redundancy (DMR) and the error is corrected by re-executing necessary operations. In this paper, based on the error recovery scheme of re-executing necessary operations, the minimization of the vote operations for error checking with respect to given resource constraints is considered. An ILP model for the optimal solution to the problem is presented and a heuristic algorithm is proposed to minimize the vote operations.
Kazuhito ITO
Saitama University
Yuto ISHIHARA
Saitama University
Shinichi NISHIZAWA
Saitama University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Kazuhito ITO, Yuto ISHIHARA, Shinichi NISHIZAWA, "Minimization of Vote Operations for Soft Error Detection in DMR Design with Error Correction by Operation Re-Execution" in IEICE TRANSACTIONS on Fundamentals,
vol. E101-A, no. 12, pp. 2271-2279, December 2018, doi: 10.1587/transfun.E101.A.2271.
Abstract: As LSI chips integrate more transistors and the operating power supply voltage decreases, LSI chips are becoming more vulnerable to the soft error caused by neutrons induced from cosmic rays. The soft error is detected by comparing the duplicated operation results in double modular redundancy (DMR) and the error is corrected by re-executing necessary operations. In this paper, based on the error recovery scheme of re-executing necessary operations, the minimization of the vote operations for error checking with respect to given resource constraints is considered. An ILP model for the optimal solution to the problem is presented and a heuristic algorithm is proposed to minimize the vote operations.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E101.A.2271/_p
Copy
@ARTICLE{e101-a_12_2271,
author={Kazuhito ITO, Yuto ISHIHARA, Shinichi NISHIZAWA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Minimization of Vote Operations for Soft Error Detection in DMR Design with Error Correction by Operation Re-Execution},
year={2018},
volume={E101-A},
number={12},
pages={2271-2279},
abstract={As LSI chips integrate more transistors and the operating power supply voltage decreases, LSI chips are becoming more vulnerable to the soft error caused by neutrons induced from cosmic rays. The soft error is detected by comparing the duplicated operation results in double modular redundancy (DMR) and the error is corrected by re-executing necessary operations. In this paper, based on the error recovery scheme of re-executing necessary operations, the minimization of the vote operations for error checking with respect to given resource constraints is considered. An ILP model for the optimal solution to the problem is presented and a heuristic algorithm is proposed to minimize the vote operations.},
keywords={},
doi={10.1587/transfun.E101.A.2271},
ISSN={1745-1337},
month={December},}
Copy
TY - JOUR
TI - Minimization of Vote Operations for Soft Error Detection in DMR Design with Error Correction by Operation Re-Execution
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2271
EP - 2279
AU - Kazuhito ITO
AU - Yuto ISHIHARA
AU - Shinichi NISHIZAWA
PY - 2018
DO - 10.1587/transfun.E101.A.2271
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E101-A
IS - 12
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - December 2018
AB - As LSI chips integrate more transistors and the operating power supply voltage decreases, LSI chips are becoming more vulnerable to the soft error caused by neutrons induced from cosmic rays. The soft error is detected by comparing the duplicated operation results in double modular redundancy (DMR) and the error is corrected by re-executing necessary operations. In this paper, based on the error recovery scheme of re-executing necessary operations, the minimization of the vote operations for error checking with respect to given resource constraints is considered. An ILP model for the optimal solution to the problem is presented and a heuristic algorithm is proposed to minimize the vote operations.
ER -