首页> 外文会议>IEEE International Verification and Security Workshop >Entropy Justification for Metastability Based Nondeterministic Random Bit Generator
【24h】

Entropy Justification for Metastability Based Nondeterministic Random Bit Generator

机译:基于非算法的无序随机比特发生器的熵理由

获取原文

摘要

We describe a minimum entropy justification for the metastable latch based nondeterministic random bit generator (NRBG) also known as an entropy source (ES). The NRBG, used for on-die generation of cryptographic keys in SOCs, is comprised of a CMOS latch with a continuously running offset cancellation loop. The offset cancellation allows for the resolution required to sample device noise at the expense of introducing serial correlation in the output data. Because the NRBG is embedded within SP 800-90 A/B/C and FIPs 140-2 compliant systems, it is critical that the loss of entropy due to serial correlation be known and bounded, and that there is a mechanism to detect loss of entropy during manufacturing test as well as normal operation. We demonstrate that a simplified one-dimensional stochastic model of the comparator in conjunction with a birth-death Markov chain model of the offset cancellation can be used to derive the minimum entropy of the NRBG and the probability of bit patterns used by entropy quality health test circuits. The result of this work compares excellently to measured data from an advanced FinFET process.
机译:我们描述了基于亚稳态的非预法随机比特发生器(NRBG)的最小熵理由,也称为熵源。用于SOC中的加密密钥的导芯生成的NRBG由具有连续运行偏移消除环路的CMOS锁存器组成。偏移消除允许以牺牲输出数据中的串行相关方式采样设备噪声所需的分辨率。由于NRBG嵌入到SP 800-90 A / B / C和FIPS 140-2兼容的系统中,因此重要的是,由于串行相关性而被众所周知和有界,熵丧失,并且有一个机制来检测损失制造测试期间的熵以及正常操作。我们证明,比较器的简化一维随机模型与偏移消除的出生死亡马尔可夫链模型可以用于导出NRBG的最小熵和熵质量健康测试所使用的比特模式的概率电路。这项工作的结果与高级FinFET过程中的测量数据卓越地比较。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号