New Mix codes for multiple bit upsets mitigation in fault-secure memories

Zhu, Ming; Xiao, Li Yi; Song, Li; Zhang, Yan Jing; Luo, Hong

doi:10.1016/j.mejo.2011.01.001

Cited by 17 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our previous research, an analysis method for the correctable probability and MTTF is proposed, which can be found in reference [5]. It is assumed that the errors that arrive at memories are assigned to a Poisson distribution.…”

Section: B Reliability Estimationmentioning

confidence: 99%

“…It is assumed that the errors that arrive at memories are assigned to a Poisson distribution. According to the reference [5], the correctable probability J of the different schemes versus radiation time is shown in Fig. 4.…”

Section: B Reliability Estimationmentioning

confidence: 99%

“…Designing EG-LDPC codes DetectorThe syndrome matrix of the (31, 16) EG-LDPC code can be obtained from equation(5).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Design EG-LDPC codes for soft error mitigation in memory

Zhu

Jing

et al. 2011

2011 Academic International Symposium on Optoelectronics and Microelectronics Technology

View full text Add to dashboard Cite

As the feature sizes of integrated circuits decreasing, single event transient (SET) in combinational circuits can not been ignored any longer. In this paper, a novel fault-secure scheme for memory has been proposed by studying the structural features of Euclidean Geometry-Low Density Parity Check (EG-LDPC) codes. The proposed fault-secure scheme can tolerate transient faults both in the storage cell and in the encoder and decoder, using the parallel majority decoding and the feedback loop structure. In order to improve the decoding speed, an algorithm is presented, which can reduce the majority decoding of EG-LDPC codes into two steps. Furthermore, the proposed scheme can suit ordinary data width (e.g., 2 n bits) in memory. Finally, the correcting capability and the reliability of the proposed scheme are analyzed. The experiment results reveal that the Mean Time to Failure (MTTF) of the proposed scheme is 419%, 104% and 118% compared with that of Hamming code, Matrix code and Reed-Muller code, respectively.

show abstract

Section: B Reliability Estimationmentioning

confidence: 99%

Section: B Reliability Estimationmentioning

confidence: 99%