Chip-scale massive-parallel ultrafast physical random bit generator

芯片级大规模并行超快物理随机位发生器

• 批准号: 1953959
• 负责人: Hui Cao
• 金额: $ 38.6万
• 依托单位: Yale University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-05-01 至 2025-04-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1953959&HistoricalAwards=false
• 关键词: Chip; scale; massive; parallel; ultrafast

The performance and reliability of our digital networked society rely on the ability to generate large quantities of randomness. Random numbers are widely used in secure communications, Monte Carlo simulations and stochastic modeling. A completely new mechanism for massive-parallel ultrafast random number generation is proposed. The random number generation rate that can be achieved with this scheme will be several orders of magnitude higher than the fastest physical random bit generators reported to date. The device is compact, robust and energy efficient. Such a parallel ultrafast random number generator is expected to have a broad impact on information security, cryptography, authentication, data integrity, and Monte Carlo simulations, etc..The ever-increasing demand to improve the security of digital information has shifted the field of random number generation from relying solely on pseudo-random algorithms to employing physical entropy sources. Ultrafast physical random number generators are key devices for achieving ultimate performance and reliability in communication and computation systems. Parallel random bit generation schemes can greatly improve the generation rate and scalability by producing many random bit streams simultaneously. The aim of this project is to develop a massive-parallel ultrafast random bit generator. It is based on a novel scheme of spatial demultiplexing of a highly multimode semiconductor laser. Ultrafast beating of numerous lasing modes produces distinct temporal fluctuations of emission intensities at different spatial locations. These random fluctuations enable parallel random bit generation in 100 to 1000 spatial channels. Each channel can produce independent random bit stream at a rate up to 1 Tb/s. The cumulative rate will approach 1 Pb/s, which is three orders of magnitude higher than the fastest physical random bit generators reported to date. The quantum mechanical noise will guarantee the true randomness of the generated bits. The preliminary studies have provided the proof-of-concept demonstration. Systematic experimental and numerical studies will be conducted to (i) understand the origin of randomness and quantify the entropy generation rate, (ii) find the fundamental limits for the number of parallel channels and the speed of random number generation, (iii) integrate photodetectors with the laser on a chip, and (iv) achieve real-time generation of random bit streams with direct interface to a computer.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

我们数字网络社会的性能和可靠性依赖于生成大量随机性的能力。随机数广泛应用于安全通信、蒙特卡罗模拟和随机建模。 提出了一种全新的大规模并行超快随机数生成机制。通过该方案可以实现的随机数生成速率将比迄今为止报道的最快物理随机位生成器高几个数量级。该设备结构紧凑、坚固且节能。 这种并行超快随机数发生器有望对信息安全、密码学、身份验证、数据完整性和蒙特卡罗模拟等产生广泛的影响。提高数字信息安全性的不断增长的需求已经转移了领域随机数生成从仅依赖伪随机算法到使用物理熵源。超快物理随机数发生器是通信和计算系统实现终极性能和可靠性的关键设备。并行随机比特生成方案可以通过同时产生许多随机比特流来极大地提高生成速率和可扩展性。该项目的目的是开发大规模并行超快随机位生成器。它基于高度多模半导体激光器的空间解复用的新颖方案。多种激光模式的超快跳动会在不同空间位置产生明显的发射强度时间波动。这些随机波动使得能够在 100 到 1000 个空间通道中并行生成随机比特。每个通道可以以高达 1 Tb/s 的速率产生独立的随机比特流。累积速率将接近 1 Pb/s，这比迄今为止报道的最快物理随机位生成器高出三个数量级。量子机械噪声将保证生成比特的真正随机性。初步研究提供了概念验证演示。 将进行系统的实验和数值研究，以（i）了解随机性的起源并量化熵生成率，（ii）找到并行通道数量和随机数生成速度的基本限制，（iii）集成光电探测器使用芯片上的激光器，以及 (iv) 通过与计算机直接接口实现随机比特流的实时生成。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优势和更广泛的影响进行评估，被认为值得支持审查标准。

项目成果

Massively parallel ultrafast random bit generation with a chip-scale laser

• DOI: 10.1126/science.abc2666
• 发表时间: 2021-02-26
• 期刊: SCIENCE
• 影响因子: 56.9
• 作者: Kim, Kyungduk;Bittner, Stefan;Cao, Hui
• 通讯作者: Cao, Hui