2021 quantum technology panorama report: China's investment exceeds 100 billion, and millions of quantum bits are put on the agenda

量子技术是一系列相关技术的未来机遇。它不仅涉及量子计算领域，而且在传感与测量、通信、仿真、高性能计算等领域具有广阔的应用前景。此外，这些量子生态系统与其整个供应链之间存在着联系。近年来，世界各国高度重视量子技术的发展，通过发布政策文件、建立研究机构、支持量子科技研究等方式加大对量子研发的投入，推动量子科技研发和产业发展，并试图在未来建立一个量子生态系统。

Quantum technology is the future opportunity of a series of related technologies. It not only involves the field of quantum computing, but also has broad application prospects in the fields of sensing and measurement, communication, simulation, high-performance computing and so on. Moreover, there are links between these quantum ecosystems and their entire supply chain. In recent years, countries around the world have attached great importance to the development of quantum technology, increased investment in quantum R & D by issuing policy documents, establishing research institutions and supporting quantum science and technology research, promoted quantum science and technology R & D and industrial development, and tried to establish a quantum ecosystem in the future.

对于本期的智能内部参考，我们推荐《光子盒2021量子技术全景展望》报告，从国家政策、最新硬件、软件和算法等方面还原量子技术的最新进展。

For the intelligent internal reference in this issue, we recommend the report "2021 panoramic prospect of quantum technology" of photon box to restore the latest progress of quantum technology from the aspects of national policies, the latest hardware, software and algorithms.

原标题：

Original title:

光子源盒

Source photon box

2021年量子技术的全景展望

Panoramic prospect of quantum technology in 2021

作者：未指明

Author: unspecified

1、 最新的国家政策和投入

1、 Latest national policies and inputs

1.欧洲联盟

1. European Union

欧洲早就认识到量子信息处理和通信技术的潜力。除了2016推出的"量子技术旗舰计划"之外，它还通过调整其他项目（如数字和空间项目）的支出来增加其可用资金，从而为实现"量子互联网"的未来愿景奠定基础。2020年5月，欧盟"欧洲量子技术旗舰计划"官方网站发布了战略研究议程（SRA）报告。

Europe has long recognized the potential of quantum information processing and communication technology. In addition to the "quantum technology flagship program" launched in 2016, it also increases its available funds by adjusting the expenditure of other programs (such as its digital and space programs), so as to lay the foundation for the realization of the future vision of "quantum Internet". In May 2020, the official website of EU's "European quantum technology flagship program" released the report of strategic research agenda (SRA).

在10年内，预计欧盟在整个量子技术旗舰计划中的相关量子支出为30-40亿欧元。

Within 10 years, it is estimated that the relevant quantum expenditure of the EU in the whole quantum technology flagship program is 3-4 billion euros.

"旗舰计划"-在扩展阶段，该计划的19个项目涵盖量子计算、通信、模拟、传感和测量以及基础科学。2020年，这些项目通过了中期审查，并启动了两个新项目——qlsi，将硅自旋量子比特添加到已成为目标的超导和离子阱量子比特中；Neasqc专门为nisq应用程序设计，以解决许多人认为缺乏软件关注的程序之间的平衡问题。

"Flagship program" - in the expansion stage, 19 projects in this program cover quantum computing, communication, simulation, sensing and measurement and basic science. In 2020, these projects passed the mid-term review and launched two new projects - qlsi to add silicon spin qubits to the ranks of superconducting and ion trap qubits that have become targets; Neasqc is designed specifically for nisq applications to address the balance of programs that many people believe lack software focus.

2.北美

2. North America

加拿大对现代量子科学做出了杰出的贡献。特别是在1984年，Gilles brassard（蒙特利尔大学）提出了著名的BB84量子密码协议。2002，量子计算研究所（IQC）在加拿大成立。2008-2018年，量子科技投资超过10亿美元。

Canada has made outstanding contributions to modern quantum science. Especially in 1984, Gilles brassard (University of Montreal) proposed the famous BB84 quantum cryptography protocol. In 2002, the Institute of quantum computing (IQC), the first in Canada, was established at the University of Waterloo. In 2008-2018, quantum science and technology investment exceeded $1 billion.

2017年，加拿大国家研究委员会（NRC）启动了一个名为quantum Canada的项目。就加拿大而言，有大量总部设在加拿大或与加拿大关系密切的知名量子公司。例如，d-wave、Xanadu、1qbit、quantum benchmark、evolutionq、Zapata和Isara。其中，创造性破坏实验室（CDL）一直是量子产业初创企业的基准。

In 2017, the National Research Council of Canada (NRC) launched a program called quantum Canada. For Canada, there are a large number of well-known quantum companies headquartered in Canada or closely related to Canada. For example, d-wave, Xanadu, 1qbit, quantum benchmark, evolutionq, Zapata and Isara. Among them, the creative destruction Laboratory (CDL) has always been the benchmark for start-ups in the quantum industry.

到2020年，加拿大量子产业将通过建立新的产业联盟巩固这一地位。2020年，温哥华的数字技术超级集群也宣布共同投资1.53亿美元。

By 2020, the Canadian quantum industry will consolidate this position by establishing a new industrial alliance. In 2020, Vancouver's digital technology super cluster also announced a co investment of $153 million.

美国在量子科学方面有着悠久的投资历史。2020年是美国国家量子计划（nqi）的第二年，随着该计划的真正形成，人们也看到了量子技术发展的亮点。Nqi将在2019-2023年花费13亿美元，并投入了大量私人资金。

The United States has a long history of investment in quantum science. 2020 is the second year of the national quantum Initiative (nqi) program of the United States, and with the real formation of the program, people also see the bright spots of the development of quantum technology. Nqi will spend US $1.3 billion in 2019-2023, and a large amount of private funds have been invested in it.

美国国家科学基金会已经建立了三个新的量子跃迁机构。这些学术领导机构将支持不同领域的研究：

Three new quantum leap institutes have been established at the National Science Foundation of the United States. These academically led institutes will support research in different fields:

1、量子科学与工程的量子感测系统（由科罗拉多大学波尔得分校领导）。量子传感技术广泛应用于精密测量；

1. Q-sense - quantum system of entanglement science and Engineering (led by the University of Colorado at Boulder). Quantum sensing technology is widely used in precision measurement;

2、Hqan -混合量子体系和网络（由香槟伊利诺伊大学领导）将开发一个用于离子阱、中性原子和超导量子比特系统的多节点测试平台，以及分布式量子计算软件栈。它还致力于下一代容错量子比特，并与芝加哥量子交易所密切合作；

2. Hqan - hybrid quantum architecture and network (led by the University of Illinois at champagne) will develop a multi node test-bed for ion traps, neutral atoms and superconducting qubit systems, as well as a distributed quantum computing software stack. It is also committed to the next generation of fault-tolerant qubits and works closely with the Chicago quantum exchange;

3、Pfqc -当前和未来量子计算（由加利福尼亚大学，伯克利领导）。设计大规模量子计算机，为当前和未来的量子计算平台开发有效的算法，并验证量子计算机能够超越经典计算机。

3. Pfqc - current and future quantum computing (led by the University of California, Berkeley). Design large-scale quantum computers, develop effective algorithms for current and future quantum computing platforms, and verify that quantum computers can surpass classical computers.

美国能源部拥有由17个国家实验室组成的独特网络，在美国研究领域具有独特的能力。美国能源部建立了五个国家量子信息科学（QIS）研究中心：

The U.S. Department of energy has a unique network of 17 national laboratories with unique capabilities in the field of U.S. research. The U.S. Department of energy has established five national Quantum Information Science (QIS) research centers:

1.Q-next-下一代量子科学与工程（阿贡国家实验室）。它将专注于远程量子网络、量子传感、处理和测试。它将建立两个用于材料和设备制造的国家量子铸造厂。著名的合作伙伴包括英特尔、IBM、微软和coldquanta。

1. Q-next - next generation quantum science and Engineering (Argonne National Laboratory). It will focus on long-distance quantum networks, quantum enabled sensing, processing and testing. It will establish two national quantum foundries for material and device manufacturing. Famous partners include Intel, IBM, Microsoft and coldquanta.

2.C2qa——量子优势协同设计中心（布鲁克海文国家实验室）。它旨在克服早期nisq设备的局限性，实现高能、核、化学和凝聚态物理科学应用中的量子优势。五年目标是在软件优化、基本材料和设备特性以及量子误差校正方面提高10倍。著名的合作伙伴包括IBM。

2. C2qa - quantum advantage Collaborative Design Center (Brookhaven National Laboratory). It aims to overcome the limitations of early nisq equipment to realize quantum advantages in high-energy, nuclear, chemical and condensed matter physical science applications. The five-year goal is to improve 10 times in software optimization, basic material and equipment characteristics and quantum error correction. Famous partners include IBM.

3.SQMS-超导量子材料和系统中心（费米国家加速器实验室）。通过理解导致退相干的物理过程，我们专注于创造更好的超导量子位。其目的是利用下一代超导量子比特技术构建量子计算机。

3. SQMS - superconducting quantum materials and Systems Center (Fermi National Accelerator Laboratory). By understanding the physical process causing decoherence, we focus on creating better superconducting qubits. The purpose is to build a quantum computer using the next generation superconducting qubit technology.

4.QSA-量子系统加速器（劳伦斯伯克利国家实验室）。它旨在联合设计算法、设备和工程解决方案，以在科学应用中提供认证的量子优势。关键技术包括中性原子、离子阱和超导量子位。桑迪亚国家实验室是主要合作伙伴。

4. QSA - quantum system accelerator (Lawrence Berkeley National Laboratory). It aims to jointly design the algorithms, equipment and engineering solutions required to provide the quantum advantage of authentication in scientific applications. Key technologies include neutral atoms, ion traps and superconducting qubits. Sandia National Laboratory is the main partner.

5.QSC-量子科学中心（橡树岭国家实验室）。利用拓扑系统算法和测量异常微弱信号的新量子系统，发现、设计和演示拓扑量子材料。微软是五大核心成员之一。其他合作伙伴包括IBM和coldquanta。

5. QSC - Quantum Science Center (Oak Ridge National Laboratory). Discovery, design and demonstration of topological quantum materials, using the algorithm of topological system and a new quantum system for measuring abnormally weak signals. Microsoft is one of the five core members. Other partners include IBM and coldquanta.

美国能源部推出量子战略后，美国国防部根据美国能源部17个国家实验室的初始骨干网络，提出了量子互联网的战略蓝图。

After the U.S. Department of energy launched the quantum strategy, the U.S. Department of defense put forward the strategic blueprint of quantum Internet according to the initial backbone network of 17 national laboratories of the U.S. Department of energy.

3.中国

3. China

中国的"五年计划"（特别是2006年以来，包括量子科学）一直在推动科学技术的发展。中国在世界中央和省级量子科技研究中心投资15亿美元。到目前为止，中国已经部署了世界上最大的QKD网络，并在先进的空间量子通信技术方面继续保持世界领先地位。"墨子卫星和九掌量子处理器是该计划成功的标志。

China's "Five-Year Plan" (especially since 2006, including quantum science) has been promoting the development of science and technology. China has invested US $1.5 billion in central and provincial quantum science and technology research centers in the world. So far, China has the world's largest deployed QKD network and continues to maintain a world leading position in advanced space quantum communication technology" Mozi satellite and Jiuzhang quantum processor are the symbols of the success of the program.

从2006年到2020年，中国计划从中央政府支出10亿美元，从地方政府支出5亿美元。据国家媒体报道，到2022年，投资将达到近150亿美元（1000亿元人民币）。目前，量子信息科学国家实验室（nlqis）的网络正在建立。

From 2006 to 2020, China plans to spend $1 billion from the central government and $500 million from local governments. According to state media reports, the investment will reach nearly 15 billion US dollars (100 billion yuan) by 2022. At present, the network of the National Laboratory for Quantum Information Science (nlqis) is being established.

1、Nlqis合肥：将成为全球最大的量子研究机构和项目总部。我们将专注于光子、钻石NV色心和硅自旋量子比特技术，以及量子通信和量子传感。

1. Nlqis Hefei: it will become the world's largest quantum research institution and the headquarters of the program. We will focus on photon, diamond NV color center and silicon spin quantum bit technology, as well as quantum communication and quantum sensing.

2.Nlqis北京：该分支机构将专注于理论、离子阱和拓扑量子比特。

2. Nlqis Beijing: this branch will focus on theory, ion traps and topological qubits.

3.Nlqis上海分公司：该分公司将专注于超导量子比特、超冷原子和自由空间量子通信。

3. Nlqis Shanghai: this branch will focus on superconducting qubits and ultracold atoms and free space quantum communication.

阿里巴巴、百度、腾讯和华为都对量子技术进行了量子投资。"量子科技"和"量子科技"计划将在2021年正式采用，其中包括"量子科技"和"量子科技"的概念，以减少对外资的依赖。与此同时，创新也是一个关键主题。

Alibaba, Baidu, Tencent and Huawei have all made quantum investments in quantum technology. GuoDun quantum and Benyuan quantum are famous start-ups" The 14th five year plan details the plan, which will be officially adopted in 2021. A key concept is the "double cycle", which includes reducing dependence on foreign high technology and increasing openness to foreign investment. At the same time, innovation is also a key theme.

2、 2021量子硬件展望

2、 2021 quantum hardware outlook

1.超导量子比特

1. Superconducting qubit

谷歌：今年硬件行业的第一条主要新闻是，约翰·马丁尼离开谷歌是因为与他的领导哈特穆特·内文关系紧张。在谷歌的量子夏季研讨会上，内文再次强调了谷歌计划的连续性，并概述了他们计划的里程碑，即到2029年用100万物理超导量子比特建造一个"小型"ftqc。

Google: the first major news in the hardware industry this year is that John martinis left Google because of tensions with his leadership, Hartmut neven. At Google's quantum Summer Seminar, neven once again emphasized the continuity of Google's plan and outlined the milestone of their plan to build a "small" ftqc with 1 million physical superconducting qubits by 2029.

即使在他卸任时，马丁尼斯也一直在强调谷歌在程序和硬件方面的领先优势。然而，也会有挑战。谷歌首选的可调谐量子比特和快速逻辑门提供了极大的灵活性和性能，但sycamore53q设备的校准显然是一个挑战。

Even when he left office, Martinis has been emphasizing Google's leading edge in program and hardware. However, there will be challenges. Google's preferred tunable qubit and fast logic gate provide great flexibility and performance, but the calibration of sycamore53q device is obviously a challenge.

有了额外的控制，额外的控制线需要在芯片上和芯片外布线。缩放会自动增加路由挑战以及组件数量与总体故障率之间的关系。值得注意的是，谷歌在2020年报告的大部分工作都使用了sycamore的23q配置，因为自动校准最初无法在大设置下提供可接受的2q门性能。谷歌将材料研究作为提高量子比特相干时间的一种方法。尽管前景很好，但这需要科学进步，而不仅仅是工程进步。

With additional control, additional control lines need to be routed on and off the chip. Scaling automatically increases the routing challenge and the relationship between the number of components and the overall failure rate. It is worth noting that most of the work reported by Google in 2020 used sycamore's 23q configuration, because automatic calibration initially failed to provide acceptable 2q gate performance in large settings. Google takes material research as a way to improve the coherence time of qubits. Although the prospect is good, it requires scientific progress, not just engineering progress.

谷歌路线图——从现在到2029年：102q（逻辑量子位原型）、103q（一个逻辑量子位）、104q（可平铺逻辑模块）、105q（工程扩展）、106q（纠错量子计算机）。通过表面代码协议进行错误纠正。

Google roadmap - from now to 2029: 102q (logic qubit prototype), 103q (one logic qubit), 104q (tilable logic module), 105q (Engineering expansion), 106q (error correction quantum computer). Error correction through surface code protocol.

IBM显然专注于大规模ftqc。Condor的原始设计使用了与其他近期芯片相同的六边形布局。这种低连接性设计是为了使具有固定频率量子位设计的芯片更易于制造，并旨在使用低连接性色码代替表面码进行纠错。此外，IBM的超级冰箱最终可以堆叠多个芯片，提供"数百万"内部连接的量子位。

IBM is clearly focused on large-scale ftqc. Condor's original design used the same hexagonal layout as other recent chips. This low connectivity design is to make the chip with fixed frequency qubit design easier to manufacture, and aims to use low connectivity color code instead of surface code for error correction. To ensure that their roadmap is clearer than other companies in 2023. In addition, IBM's super refrigerator can eventually stack multiple chips, providing "millions" of internally connected qubits.

在评估IBM是否能够实现其目标时，很难不被其过去的记录所打动。IBM还将显著减少第2季度门错误。尽管最近几代处理器的关键参数有了显著的改进，但现在需要认识到这一点。

When assessing whether IBM can achieve its goals, it is difficult not to be impressed by its past record. IBM will also significantly reduce 2q gate errors. Although their recent generations of processors have shown stable improvements in this key parameter, their plans now seem to recognize the need for more significant changes to their 2q gate design.

IBM一直试图在每个门中使用额外的可调谐振荡器耦合器和旁路电容耦合器，同时保留固定频率的量子位，以利用其允许的长相位干扰时间。这确保了2q门速度更快（误差更低），但由于与以前的技术设计相比有重大变化，目前只能在简单的2q实验设备中实现。

IBM has been trying to use additional tunable oscillator couplers and bypass capacitor couplers in each gate while retaining fixed frequency qubits to take advantage of their allowed long phase interference time. This ensures that the 2q gate is faster (with lower error), but due to significant changes compared with its previous technical design, it can only be realized in a simple 2q experimental equipment so far.

为了应对日益增长的布线挑战，IBM开发了基于三层超导布线的下一代芯片布局。看到这些技术如何顺利地结合在一起将是对IBM路线图的一个关键测试。

In order to meet the growing cabling challenges, IBM has developed a next-generation chip layout based on three-layer superconducting cabling. Seeing how these technologies fit together smoothly will be a key test of IBM's roadmap.

在IqT欧洲峰会上，Lieven vandersypen（qutech的科学总监）指出，尽管许多人希望更快的发展，但将量子量逐年翻倍仍然是一个挑战。我们需要记住，我们不能仅仅通过添加量子位来实现这一点。同步操作中2q门的保真度是电流限制因素。Jay Gambetta（IBM）表示："我看到了未来的挑战，但没有障碍。"

At the IqT European summit, Lieven vandersypen (scientific director of qutech) pointed out that although many people want faster development, doubling the quantum volume year by year is still a challenge. We need to recall that we can't do this just by adding qubits. The fidelity of 2q gate in synchronous operation is the current limiting factor. Jay Gambetta (IBM) said, "I see the challenges ahead, but there are no obstacles."

2.离子阱

2. Ion trap

离子阱的开发在2020年非常顺利。霍尼韦尔已经成为第一家通过其6q H0和10q H1处理器实现QV 64和QV 128的制造商。有些人可能会想，10q处理器怎么能声称自己的功能和IBM的27q处理器一样强大？然而，这仅仅突出了离子阱研究人员长期以来描述的两个优点：与超导量子比特方法相比，它具有更高的连接性和更高的门保真度。这两个优点可以确保更高的QV。霍尼韦尔处理器也是第一个实现中间电路测量的处理器，进一步提高了灵活性。

The development of ion trap is very smooth in 2020. Honeywell has become the first manufacturer to achieve QV 64 and QV 128 with its 6q H0 and 10q H1 processors. Some people may wonder, how can the 10q processor claim to be as powerful as IBM's 27q Processor? However, this just highlights two advantages that ion trap researchers have described for a long time: compared with superconducting qubit method, it has superior connectivity and higher gate fidelity. These two advantages can ensure higher QV. Honeywell processor is also the first processor to realize intermediate circuit measurement, which further improves the flexibility.

霍尼韦尔路线图（不同量子比特布局）-2020-2030年，H1（线性离子阱），H2（跑道布局），H3（网格布局），H4（集成光学元件），H5（大规模瓷砖）。

Honeywell roadmap (different qubit layouts) - 2020-2030, H1 (linear ion trap), H2 (runway layout), H3 (grid layout), H4 (integrated optical elements), H5 (large-scale tiling).

离子阱结构通常使用模块之间的光子互连进行扩展。最近已经证明了更快的互连，但它似乎仍然是性能瓶颈。另一方面，宇宙量子已经证明，他们的离子穿梭方法可以提供类似于全连接的QV。

Ion trap architectures are usually extended using photonic interconnects between modules. Faster interconnection has been demonstrated recently, but it still seems to be a performance bottleneck. On the other hand, universal quantum has proved that their ion shuttle method can provide QV similar to full connection in principle.

3.中性原子

3. Neutral atom

到2020年，中性原子量子位将继续突飞猛进地发展。它们具有许多与离子阱相同的特性。它们的优点是中性原子可以被更紧密地包裹。这意味着它可以更快地扩展到1000q模块。这项技术也被称为冷原子，因为它利用激光冷却和高真空达到毫开尔文（MK）的温度，远低于低温冷却的范围。

Neutral atom qubits will continue to develop by leaps and bounds in 2020. They have many of the same characteristics as ion traps. Their advantage is that neutral atoms can be wrapped more closely. This means that it can be extended to 1000q module faster. This technology is also called cold atom because it uses laser cooling and high vacuum to reach the temperature of millikelvin (MK), which is far lower than the range of low-temperature cooling.

Coldquanta是一家采用这种方法的知名公司，它已将quantum core作为一个基本单元推出，以瞄准许多量子领域的机遇。它也是阿尔伯特的基础，阿尔伯特是云上的量子物质系统。冷量子已经被发现

Coldquanta, a well-known company adopting this approach, has launched quantum core as a basic unit to target opportunities in many quantum fields. It is also the basis of Albert, the quantum matter system on the cloud. Coldquanta has been

DARPA（国防高级研究计划局）被选中参与作为onisq项目一部分的1000q处理器的开发。该奖项的价值高达740万美元。

DARPA (Defense Advanced Research Projects Agency) was selected to participate in the development of 1000q processor as part of onisq program. The value of this award is as high as US $7.4 million.

Coldquanta路线图——2021年达到100个，2022年达到300个，2024年达到1000个。其他选择中性原子的公司包括quera、paswal和atom computing。

Coldquanta roadmap - 100Q by 2021, 300q by 2022 and 1000q by 2024. Other companies that choose neutral atoms include quera, paswal and atom computing.

4.硅量子点

4. Silicon quantum dots

2020年，基于量子点的硅量子比特在实现其长期承诺的优势方面取得了重大进展。QuTeo和新南威尔士大学（UNSW）在1K的温度下用金属氧化物半导体（MOS）量子点演示量子位运算。尽管相干时间和保真度在这些较高温度下是否具有竞争力仍有待观察，但预计这将是一种更容易操作和扩大设备规模的机制。

In 2020, quantum dot based silicon qubits have made significant progress in realizing one of the advantages of their long-term commitments. Qutech and the University of New South Wales (UNSW) demonstrated qubit operation with metal oxide semiconductor (MOS) quantum dots at a temperature of 1K. This is expected to be a significantly easier mechanism to operate and expand the scale of the equipment, although it remains to be seen whether the coherence time and fidelity are competitive at these higher temperatures.

澳大利亚初创企业硅量子计算公司一直是硅量子比特的早期推动者。2020年，它宣布了路线图的重点，放弃了MOS量子点，增加了磷原子量子位。这些设备使用超尖端制造技术，提供超越传统CMOS技术的原子精度方法。

Silicon quantum computing, an Australian start-up, has always been an early promoter of silicon qubits. In 2020, it announced the focus of its roadmap, abandoned MOS quantum dots and added phosphorus atom qubits. These devices use ultra cutting-edge manufacturing technology to provide atomic precision methods that go beyond traditional CMOS technology.

在描述SQC的制造技术时，Michelle Simmons（SQC的创始人）指出，不仅可以以原子精度设计量子比特，而且同样的技术可以在相同的设备基板上创建稳定、简单和原始的控制电路。今年，他们报告说硅量子比特达到了迄今为止最低的噪声。离开谷歌后，约翰·马丁尼（John martinis）现在加入了SQC，并将开发具有超快门和可扩展布线选项的设备。

When describing the manufacturing technology of SQC, Michelle Simmons (founder of SQC) pointed out that not only can qubits be designed with atomic precision, but also the same technology can create stable, simple and original control circuits in the same device substrate. This year, they reported that silicon qubits achieve the lowest noise to date. After leaving Google, John martinis has now joined SQC and will develop devices with ultra fast doors and scalable cabling options.

2020年，加拿大初创公司光子公司（photon Inc.）发表了早期研究报告，承诺为硅量子比特"军火库"添加一个重要的新工具。这有望改善基于硅中t中心缺陷的光子界面。

In 2020, Canadian start-up photon Inc published early research, promising to add an important new tool to the silicon qubit "arsenal". This is expected to improve the interface of photons based on t-centre defects in silicon.

5.光量子比特

5. Light quantum bit

在中国进行的九章实验可以证明，这种计算比目前在任何其他平台上实现的计算都要复杂。在第九章中，我们实现了一种称为高斯玻色采样的算法来实现这一点。成功地构建了一个包含76个光子和100个模式的高斯玻色取样量子计算原型。在200秒内生成的输出样本表明，世界上最强大的超级计算机——福库需要6亿年才能实现。这比谷歌最初展示的梧桐树的量子优势要复杂得多。

The nine chapter experiment in China can prove that this calculation is more complex than that implemented on any other platform so far. In Chapter 9, an algorithm called Gaussian Bose sampling is implemented to achieve this. A Gaussian Bose sampling quantum computing prototype with 76 photons and 100 modes is successfully constructed. The output samples generated in 200 seconds claim that fugaku, the world's most powerful supercomputer, will take 600 million years to realize. It is much more complex than Google's initial demonstration of the quantum advantage of sycamore.

第九章并非空穴来风。至少自2006年以来，中国一直在增加对量子技术的投资。潘建伟团队的专业知识是众所周知的。2019年，他们首次实现了20个光子和60条模式干涉线的玻色取样量子计算。最新的实验是一项了不起的科学成就，也是一项杰作，再次证明了该团队的科学和工程技术。

Chapter nine does not come out of thin air. China has been increasing its investment in quantum technology since at least 2006. The expertise of Pan Jianwei's team is well known. In 2019, they realized the Bose sampling quantum calculation of 20 photons and 60 mode interference lines for the first time. The latest experiment is a remarkable scientific achievement and a masterpiece that once again proves the team's science and engineering technology.

第九章的含义不需要细分。设备的当前形式不可编程。它实现了一种静态算法，而不是一般的量子计算方法。也许更重要的是，它是通过"传统"光学平台安装实现的。所有活动组件仍然是离散的。为了实现稳定的配置，需要进行许多手动调整。这种方法在科学上是令人兴奋的，但它对扩大规模提出了严峻的挑战。

The meaning of chapter nine does not need to be broken down. The current form of the device is not programmable. It implements a static algorithm rather than a general quantum computing method. Perhaps more importantly, it is achieved through "traditional" optical platform installation. All active components are still discrete. In order to achieve a stable configuration, many manual adjustments are required. This method is exciting in science, but it poses a severe challenge to expand the scale.

6.未来展望

6. Future outlook

量子霸权——不管关于计算难度的辩论最终结果如何，第九章都会引起轩然大波。注意该技术是否可编程和可扩展。只有这样，它才能真正具有颠覆性。

Quantum hegemony - regardless of the final outcome of the debate on the difficulty of calculation, Chapter 9 will cause an uproar. Pay attention to whether this technology can be programmable and scalable. Only in this way can it be truly subversive.

QV-ionq对其新的32q量子计算机有很高的期望。测量的性能真的会达到400万QV（在新的ionq术语中为22aq）吗？

QV - ionq has high expectations for its new 32q quantum computer. Will the measured performance really reach 4 million QV（ (22aq in the new term of ionq)?

量子体积可以用来测量许多性能指标，但它不能测量不同量子比特平台之间的初始门速度差。预计超导量子比特团体将采取适当措施进行反击，以便他们能够展示自己的快门。

QV / S - quantum volume can be used to measure many performance indicators, but it can not measure the difference of original gate speed between different quantum bit platforms. It is expected that the superconducting qubit community will take appropriate measures to fight back so that they can show their fast gate.

量子位的数量——IBM会率先在云中部署100Q+处理器和127QEagle吗？还是里格蒂会使用4x32q多芯片Aspen模块抢占先机？我们将从谷歌的"100Q"设备中看到什么？注意同步2q门保真度的趋势。

Number of qubits - will IBM be the first to put 100Q + processors and 127q eagle in the cloud? Or will rigetti use 4x32q multichip Aspen modules to seize and take the lead? What will we see from Google's "100Q" device? Note the trend of synchronous 2q gate fidelity.

逻辑量子位——关注主要参与者的纠错演示，这表明他们正在迈向一个新的重要里程碑——操作逻辑量子位。

Logical qubits - focus on the error correction demonstration of major players, which shows that they are moving towards a new important milestone - operating logical qubits.

中国-本源量子能为云产品添加60量子位的婺源2.0设备吗？

China - can Benyuan quantum add 60 qubits of Wuyuan 2.0 devices to cloud products?

欧洲-QT旗舰项目opensuperq预计将交付其第一台设备。距离100Q有多远？对应的QV是多少？Aqion将交付50个以上的设备；注意其基于机架的灵活配置。

Europe - QT flagship project opensuperq is expected to deliver its first equipment. How far is it from 100Q and what is the corresponding QV? Aqion will deliver 50Q + equipment; Note its rack based flexible configuration.

英国-里格蒂正在英国建造一台基于量子比特的超导机器，该机器位于牛津仪器公司最新的proteox系列稀释冰箱中。请注意这个量子计算机可能发布的细节。

UK - rigetti is building a superconducting qubit based machine in the UK, which is housed in the latest proteox series dilution refrigerator of Oxford Instruments. Pay attention to the details that may be released by this quantum computer.

超导技术——观察初创企业的发展和该技术的新特点，如seeqc和OQC。特别注意QCI计划的细节。这是一支强大的团队，但仍在暗中运作。他们选择支持的内容将成为他们如何看待未来扩张挑战的指南。

Superconducting technology - observe the development of start-ups and the new features of this technology, such as seeqc and OQC. Pay special attention to the details of QCI plan. This is a strong team, but it is still operating in the dark. What they choose to support will be a guide to how they see the challenges of future expansion.

量子退火——关注未来d波硬件计划的细节。七里马扎罗提出的"相干"量子退火机的细节是什么。

Quantum annealing -- pay attention to the details of the future d-wave hardware plan. What are the details of the "coherent" quantum annealer proposed by qilimanjaro.

离子阱技术——霍尼韦尔和ionq等现有领先企业将由AQT推动。关注牛津ionics、universal quantum和nextgenq等微波门技术初创公司的新闻。

Ion trap technology - existing leaders such as Honeywell and ionq will be driven by AQT. Follow news from microwave gate technology startups such as Oxford ionics, universal quantum and nextgenq.

中性原子——2021年我们会看到coldquanta的100Q装置成为头条新闻吗？要特别注意忠实。关注更多来自创业公司quera、Pascal和atom computing的项目。

Neutral atom - will we see coldquanta's 100Q device make the headlines in 2021? Pay particular attention to fidelity. Follow more programs from startups quera, Pascal and atom computing.

量子点-自旋量子比特原型通常基于金属氧化物半导体或硅衬底上的SiGe量子点。在过去两年中，硅衬底上的锗量子比特取得了惊人的进步，包括qutech演示的4Q处理器。观察哪种变体将成为领先的量子点量子位平台。

Quantum dots - spin qubit prototypes are usually based on metal oxide semiconductors or SiGe quantum dots on silicon substrates. In the past two years, germanium qubits on silicon substrates have made amazing progress, including the 4Q processor demonstrated by qutech. Observe which variant will become the leading quantum dot quantum bit platform.

光子平台——这项技术的许多变体已经开发出来，每种都有不同的优缺点。绝缘体上硅（SOI）是最成熟的技术，由psiq支持。氮化硅（Si3N4）提供了强大的现有组件生态系统，受到Xanadu和quix的青睐。这项技术的其他变化正在出现。请注意这些技术中哪些将在量子应用中脱颖而出。二元性将选择哪一个作为它的起点？

Photon platform - many variants of this technology have been developed, each with different advantages and disadvantages. Silicon on insulator (SOI) is the most mature technology and is supported by psiq. Silicon nitride (Si3N4) provides a powerful existing component ecosystem and is favored by Xanadu and quix. Other changes in this technology are emerging. Note which of these technologies will stand out in quantum applications. Which one will duality choose as its starting point?

拓扑量子位——在2020年经历了一次重大挫折，质疑了TU Delft之前关于mayorana准粒子的结果。继续关注相关争议。

Topological qubits - experienced a major setback in 2020, questioning the previous results of TU Delft on mayorana quasiparticles. Continue to pay attention to relevant disputes.

英国国家量子计算中心（nqcc）在2021年初专注于资助许多技术领域的项目。

The national Quantum Computing Center (nqcc) of the UK focuses on funding programs in many technical fields in early 2021.

3、 量子软件展望

3、 Prospect of quantum software

IBM继续领导量子云平台的研发。这将是未来量子计算发展的关键。它可以处理更复杂的操作，如果量子计算服务在竞争日益激烈的云计算服务市场中推出，它将占据绝对优势。从长远来看，支持开发者是传统软件行业的任务，但我们需要记住，量子计算的底层技术与软件行业的本质不同。我们应该对研发有彻底的革命精神。

IBM continues to lead the research and development of quantum cloud platform. This will be the key to the development of quantum computing in the future. It can handle more complex operations, and if quantum computing services are launched in the increasingly competitive cloud computing service market, it will occupy an absolute advantage. In the long run, supporting developers is the task of the traditional software industry, but we need to keep in mind that the underlying technology of quantum computing is different from the essence of the software industry. We should have a thorough revolutionary spirit towards R & D.

在数字革命中，许多人认为它在新的软件领域也至关重要。不同的参与者正在研究不同的策略。今天，早期的量子社区和生态系统已经形成。

In the digital revolution, many people believe that it is also crucial in the new field of software. Various participants are studying different strategies. Today, early quantum communities and ecosystems have taken shape.

1.量子计算云服务

1. Quantum computing cloud service

IBM问：IBM宣布了其量子计算研发领域的一个新里程碑。在过去的四年中，IBM云上部署了28个量子计算系统，其中8个系统的量子容量为32个。IBM Q网络有115个客户、政府、初创企业、合作伙伴和大学成员。IBM quantum experience拥有超过250000名注册用户，超过10亿用户通过IBM云定期运行IBM quantum系统中的电路。研究人员使用IBM quantum systems发表了250多篇学术论文。IBM的量子计算机服务IBM Q for business已取得阶段性成功。

IBM Q: IBM announced a new milestone in its quantum computing R & D territory. In the past four years, 28 quantum computing systems have been deployed on IBM cloud, of which 8 systems have a quantum volume of 32. IBM Q network has 115 customers, governments, start-ups, partners and university members. IBM quantum experience has more than 250000 registered users, and more than 1 billion users regularly run circuits in IBM quantum systems through IBM cloud. Researchers have published more than 250 academic papers using IBM quantum systems. IBM's quantum computer service IBM Q for business has achieved phased success.

当其他竞争对手开始建立自己的quantum社区时，IBM根据其早期建立quantum社区的经验出版了一些书籍。他们指出，在不断发展的供应商生态系统中，可以提供与qiskit兼容的库和工具，而不仅仅是IBM硬件。

When other competitors began to build their own quantum community, IBM published books based on its experience in building quantum community at an early stage. They point out that in the evolving vendor ecosystem, libraries and tools compatible with qiskit can be provided, not just IBM hardware.

D-wave于2018年10月推出leap云平台，提供基于D-wave量子退火处理器的量子计算云服务。

D-wave launched leap cloud platform in October 2018 to provide quantum computing cloud services based on d-wave quantum annealing processor.

量子计算的先驱Rigetti computing推出了Rigetti量子云服务（QCS）——一个使用Rigetti的混合量子/经典方法开发和运行量子算法的完整平台。量子优势是利用量子计算技术解决重要或有价值的商业问题。最近，越来越多实力强大的量子公司开始投资量子云服务平台的研发。亚马逊AWS发布了量子计算服务braket。此外，AWS将推出"AWS量子计算中心"和"亚马逊量子解决方案实验室"，以促进量子计算领域的更多合作。

Rigetti computing, a pioneer in quantum computing, launched rigetti quantum cloud service (QCS) - a complete platform for developing and running quantum algorithms using rigetti's hybrid quantum / classical methods. Quantum advantage is to use quantum computing technology to solve important or valuable business problems. Recently, more and more powerful quantum companies have begun to invest in the research and development of quantum cloud service platform. Amazon AWS releases braket, a quantum computing service. In addition, AWS will launch "AWS quantum computing center" and "Amazon quantum solutions laboratory" to promote more cooperation in quantum computing.

2.高性能模拟器

2. High performance simulator

对于量子发展来说，高性能仿真是一个关键环节。随着被模拟量子比特数的增加，量子模拟器的发展迫在眉睫。IBM quantum支持一系列离线和在线模拟器。谷歌的高性能开源量子电路模拟器QSIM已经证明，它可以在111秒内在谷歌云节点上模拟14门深度的32量子位量子电路。Amazon braket和azure quantum等参与者非常重视他们灵活配置传统云硬件以满足用户需求的能力。Amazon braket提供了一个完全托管的高性能张量网络模拟器（TN1），它可以支持多达50个量子位的量子计算模拟。

For quantum development, high-performance simulation is a key link. With the increase of the number of qubits to be simulated, the development of quantum simulator is imminent. IBM quantum supports a range of offline and online simulators. Google's high-performance open source quantum circuit simulator QSIM has proved that it can simulate a 32 qubit quantum circuit with 14 gate depth in a Google cloud node in 111 seconds. Participants like Amazon braket and azure quantum attach great importance to their ability to flexibly configure traditional cloud hardware to meet user needs. Amazon braket provides a fully hosted high-performance tensor network simulator (TN1), which can support quantum computing simulation of up to 50 qubits.

Atos是数字转换领域的全球领导者，也是第一家成功模拟量子噪声的公司。其量子模拟器ATOS量子机器学习机（ATOS QLM）被称为世界上最好的商用量子模拟器

Atos is a global leader in digital transformation and the first company to successfully simulate quantum noise. Its quantum simulator ATOS quantum machine learning machine (ATOS QLM) is known as the best commercial quantum simulator in the world

该模拟器将高功率超小型机器与通用编程语言相结合，使研究人员和工程师能够开发和测试量子软件。Atos在许多国家安装了量子学习机，包括奥地利、丹麦、法国、德国、荷兰和美国。量子模拟器可以模拟多达40个量子位。

The simulator combines high-power and ultra compact machines with general programming language, enabling researchers and engineers to develop and test quantum software. Atos has installed quantum learning machines in many countries, including Austria, Denmark, France, Germany, the Netherlands and the United States. The quantum simulator can simulate up to 40 qubits.

中国云产品目前强调量子计算模拟器的开发。华为的HiQ 2.0（仅因监管原因在亚洲使用）最多可以模拟42个量子位。阿里巴巴的ac qdp声称，它可以在某些应用程序中使用，即使是50量子位。Source quantum最近通过访问其一个6位量子处理器（计划扩展到24位，正在进行）推出了一个基于真实量子计算机的云。

Chinese cloud products currently emphasize the development of quantum computing simulators. Huawei's HiQ 2.0 (used only in Asia for regulatory reasons) can simulate up to 42 qubits. Alibaba's ac-qdp claims that it can be used in some applications even at 50 qubits. Source quantum recently launched a cloud based on real quantum computers by accessing one of its 6-bit quantum processors (planned to expand to 24 bits, ongoing).

3.量子编译器

3. Quantum compiler

与传统编译器相比，优化量子编译器是量子研发阶段的一项重大挑战。量子计算设备在物理量子位之间的连接有限，因此双门只能应用于有限数量的量子位对。在现实世界中，量子设备中存在噪声，但是可以开发一种描述大型量子计算机噪声的算法来解决这个问题。从技术上讲，我们实际上是在谈论代码转换，因此互操作性是一个有用的特性。

Compared with traditional compilers, optimizing quantum compilers is a major challenge in the quantum R & D stage. Quantum computing devices have limited connections between physical qubits, so that double gates can only be applied to a limited number of qubit pairs. There is noise in quantum devices in the real world, but an algorithm for characterizing the noise of large quantum computers can be developed to solve this problem. Technically, we're actually talking about transcoding, so interoperability is a useful feature.

编译器市场上有几个很有前途的方向，这些方向基于深厚的专业知识，在许多情况下是互补的，而不是竞争性的。随着早期在量子硬件上实现纠错码的竞争日益激烈，编译器创新将迎来新的浪潮。

There are several promising directions in the compiler market, which are based on deep professional knowledge, which is complementary rather than competitive in many cases. With the increasingly fierce competition in the early implementation of error correction code on quantum hardware, compiler innovation will usher in a new wave.

4.量子操作系统

4. Quantum operating system

量子软件产业的前景令人印象深刻，但在量子计算机硬件迅速发展的今天，如果没有操作系统，量子计算机的实用性将大大降低。

The prospect of quantum software industry is impressive, but today, with the rapid growth of quantum computer hardware, the practicability of quantum computer will be greatly reduced without operating system.

Riverlane的三角洲流。OS是一种全新的全栈量子操作系统。总部位于剑桥的量子计算软件开发商riverlane领导的财团从英国政府获得760万英镑（约6900万人民币）的赠款，用于部署高度创新的量子操作系统deltaflow.os。

Riverlane's deltaflow. OS is a new full stack quantum operating system. The consortium led by riverlane, a Cambridge based quantum computing software developer, received a grant of £ 7.6 million (about 69 million RMB) from the British government to deploy the highly innovative quantum operating system deltaflow.os.

与其他旨在吸引早期用户的软件平台形成鲜明对比的是，deltaflow。操作系统解决了一个非常重要的问题——实现硬件和软件之间的交互，充分利用量子计算性能。为此，它提供了加速开发、低延迟以及应用程序和控制层之间灵活交互的潜力。

In sharp contrast to other software platforms designed to attract early users, deltaflow. OS solves a very important problem - realizing the interaction between hardware and software and making full use of quantum computing performance. To this end, it offers the potential for accelerated development, low latency, and flexible interaction between the application and control layers.

三角洲流。操作系统：量子处理器通常由传统的主机处理器驱动。在这两者之间，想象一个由全局和本地控制节点组成的网络。通过FPGA实现，简化了任务的获取，简化了任务的调试。这种方法有望缩短研发周期。它还使用了分布式而非分层网络节点的概念，并公开了整个量子计算堆栈的不同元素。这些函数可以最大限度地减少运行时延迟。

Deltaflow. OS: quantum processors are usually driven by conventional host processors. In between, imagine a network of global and local control nodes. Deltaflow.os simplifies the task of acquiring custom code to the control node implemented by FPGA, and emphasizes the simplified instruction set implementation, which is easier to debug. This method is expected to shorten the R & D cycle. It also uses the concept of distributed rather than hierarchical network nodes, and exposes different elements of the entire quantum computing stack. These functions are expected to minimize runtime latency.

Deltaflow.os现在发布了它的第一个版本，它与artiq（一种流行的离子阱控制系统）集成，称为"artiq上的Deltaflow"。这是该公司支持量子计算技术发展的最新里程碑，标志着riverlane建立高性能量子操作系统的目标迈出了重要一步，该系统可移植到所有量子位技术，并可扩展到数百万量子位。

Deltaflow.os has now released its first version, which is integrated with artiq (a popular ion trap control system) as "deltaflow on artiq". This is the latest milestone in the company's development of technologies supporting quantum computing, marking an important step towards riverlane's goal of building a high-performance quantum operating system that can be transplanted to all quantum bit technologies and can be extended to millions of quantum bits.

5.未来的应用

5. Future applications

真正的量子应用还为时过早，但初创公司已经在为这个未来市场定位。关键在于平衡量子算法的专业知识和深入的行业洞察力。与传统的应用软件领域相比，后者可能更为重要。建立一个能够与行业互动并蓬勃发展的商业模式是一个挑战。

Real quantum applications are still too early, but startups are already positioning for this future market. The key point is to balance the expertise of quantum algorithms with in-depth industry insight. Compared with the traditional application software field, the latter may be even more important. It is a challenge to establish a business model that can interact with the industry and flourish.

算法专家正在围绕客户开展试点合作，为特定行业建立工具和库。例如，量子化学中的1qbit qemist、CQC中的eumen和HQs中的QAD云。

Algorithm experts are conducting pilot cooperation around customers to establish tools and libraries for specific industries. For example, 1qbit qemist in quantum chemistry, eumen in CQC and QAD cloud in HQs.

其他公司强调与传统人工智能和数据科学技术的协同作用。例如，多元宇宙中的QDL和FS；药物设计中的蛋白质质量。类似地，Qu&Co强调了其与薛定谔（Schr dinger）的战略合作伙伴关系，薛定谔是当今使用的传统量子化学软件的领导者。

Other companies emphasize synergies with traditional AI and data science and technology. For example, QDL and FS in multiverse; Proteinqure in drug design. Similarly, Qu & Co emphasized its strategic partnership with Schr dinger, a leader in conventional quantum chemistry software used today.

其他人使用量子退火和量子启发式算法来获得好处。例如，1qcloud中的polarisqb和1qbit。

Others use quantum annealing and quantum heuristic algorithms to gain benefits. For example, polarisqb and 1qbit in 1qcloud.

其他人则强调他们在算法研究和开发方面的学术实力。这有助于确保政府资金用于尖端创新。例如，phasecraft和Beit。

The rest emphasize their academic strength in algorithm research and development. This helps ensure that government funds are used for cutting-edge innovation. For example, phasecraft and Beit.

4、 量子算法展望

4、 Prospect of quantum algorithms

虽然已经取得了进展，但要在当前的量子设备上证明量子算法的优势还为时过早。大规模量子计算机的未来是有希望的，但是花费是巨大的。即使有一百万个量子设备，如果使用当前的量子纠错方案，其结果也只能达到量子计算机效率的一半。

Although progress has been made, it is still too early to prove the advantages of quantum algorithms on current quantum devices. The future of large-scale quantum computers is promising, but the expenditure is large. Even with one million quantum devices, if the current quantum error correction scheme is used, it will only get half the result with twice the effort for the efficiency of quantum computers.

量子计算机可以很容易地解决世界上最强大的经典计算机需要很长时间才能解决的复杂计算问题。早期的理论工作主要集中在容错量子计算机项目的研究上。在某些情况下，例如使用shor量子算法（密码分析）、harrow Hasidim&Lloyd（线性代数）和相位估计（量子化学），运行速度可以成倍提高。

Quantum computers can easily solve the complex computing problems that the world's most powerful classical computers need a long time to solve. The early theoretical work usually focused on the research of fault-tolerant quantum computer project. In some cases, such as using shor quantum algorithm (cryptanalysis), harrow Hasidim & Lloyd (linear algebra) and phase estimation (quantum chemistry), the running speed can be increased exponentially.

然而，这种理想化的加速度只是一种假设。近年来，人们的研究主要集中在如何将量子启发式算法应用于nisq器件（如vqe、QOA、QNN和量子退火），但如何实现加速的理论基础尚未正式建立。

However, this idealized acceleration is only an assumption. In recent years, people's research has focused on how quantum heuristic algorithms are used in nisq devices (such as vqe, qaoa, QNN and quantum annealing), but the theoretical basis of how to achieve acceleration has not been officially established.

2019年底，谷歌的sycamore量子芯片取得了巨大成功，极大地提高了量子计算的性能。谷歌将"春季量子研讨会"作为其量子计算服务的亮点，量子研究成果令人印象深刻。谷歌对梧桐树的研究主要集中在如何缩短量子计算的时间并使其具有更高的计算性能。

Google's sycamore quantum chip was a great success at the end of 2019, greatly improving the performance of quantum computing. Google takes the "spring quantum seminar" as the highlight of its quantum computing service, and the quantum research results are impressive. Google's research on sycamore focuses on how to shorten the time of quantum computing and make it have higher computing performance.

采用变分量子本征值求解算法（Vqe）模拟化学反应过程——分子电子能量的Hartree-Fock计算。尽管所进行的计算也可以在经典计算机上运行，但该实验已经为量子化学模拟构建了许多关键构件，为实现化学问题的量子计算铺平了道路。

Vqe (variational quantum eigenvalue solving algorithm) is used to simulate the chemical reaction process - Hartree Fock calculation of molecular electron energy. Although the calculations performed can also run on classical computers, the experiment has built many key building blocks for quantum chemical simulation, paving the way for the realization of quantum computing for chemical problems.

Trotterization用于模拟材料科学中非常流行的8位一维费米-哈伯德模型。令人惊讶的是，谷歌成功实现量子算法所需的量子电路的深度接近500，远远超过当前设备的预期。

Trotterisation is used to simulate 8-bit 1D Fermi Hubbard model, which is very popular in material science. Surprisingly, the depth of quantum circuits required for Google to successfully implement quantum algorithms is close to 500, much deeper than expected by current devices.

在多重化学量子模拟的情况下，Google演示了基于N表示的错误缓解决策，这大大提高了实验的有效保真度。

In the case of multiple chemical quantum simulations, Google demonstrated the error mitigation decision based on N representation, which greatly improved the effective fidelity of the experiment.

从金融服务到物流再到制造业，优化算法是在实际案例中应用的重要前提。早期的优化算法包括在传统硬件上运行的Qubo、QOA和量子启发式算法。BBVA已经完成了一系列金融领域应用的初步项目，包括初创企业多元宇宙计算公司和Zapata。BBVA和多元宇宙之间的合作是动态投资组合优化的经典案例。该组合现在已用于评估各种早期量子硬件，包括nisq测试、量子退火解决方案和量子启发式算法。BBVA的结果表明，量子退火解和量子启发式算法能够很好地解决投资组合问题。

From financial services to logistics to manufacturing, optimization algorithm is an important premise for application in practical cases. Early optimization algorithms include Qubo, qaoa and quantum heuristic algorithm running on traditional hardware. BBVA has completed a series of preliminary projects for applications in the financial field, including startups multiverse computing and Zapata. The cooperation between BBVA and multiverse is a classic case of dynamic portfolio optimization. The portfolio has now been used in the evaluation of various early quantum hardware, including the testing of nisq, quantum annealing solution and quantum heuristic algorithm. The results of BBVA show that quantum annealing solution and quantum heuristic algorithm can solve the portfolio problem well.

在欧洲，汽车行业尤其活跃。大众、宝马和零件供应商博世等公司去年讨论了他们的经验。优化算法再次被视为物流和制造运营中的关键机会。

In Europe, the automotive industry is particularly active. Companies such as Volkswagen, BMW and parts supplier Bosch discussed their experiences last year. Optimization algorithms are again seen as key opportunities in logistics and manufacturing operations.

5、 量子互联网展望

5、 Prospect of quantum Internet

量子计算机将对商业和社会产生许多积极影响，但现在世界还不知道。当量子计算机升级到一定程度时，它将能够打破当前网络安全所依赖的公钥加密系统。更糟糕的是，量子技术可以解密当前截获和存储的各种数据。

Quantum computer will have many positive effects on business and society, but it is not known to the world now. When the quantum computer is upgraded to a certain extent, it will be able to break the public key encryption system that the current network security depends on. Worse, quantum technology can decrypt all kinds of data currently intercepted and stored.

目前，学术研究正在迅速跟进，以填补技术发展带来的新漏洞。它包括基于数学原理的后量子密码学（PQC）和基于物理原理的量子密码学方法，特别是量子随机数发生器（QRNG）和量子密钥分配（QKD）。

At present, academic research is rapidly following up to fill the new loopholes brought by technological development. It includes post quantum cryptography (PQC) based on mathematical principles and quantum cryptography methods based on physical principles, especially quantum random number generator (QRNG) and quantum key distribution (QKD).

然而，要了解这个快速发展的行业，我们还必须研究一种更重要的未来技术形式——量子互联网。

However, to understand this rapidly developing industry, we must also study a more important future technology form - Quantum Internet.

今天，许多著名的量子硬件公司已经制定了未来十年内生产一百万量子比特处理器的路线图。严格地说，这些设备可能仍然不足以打破我们目前的互联网标准（估计8小时内有2000万量子比特）。此外，这些计划不可能一帆风顺。

Today, many well-known quantum hardware companies have formulated a roadmap for the future of producing a million qubit processor within ten years. Strictly speaking, such devices may still not be enough to break our current Internet standards (an estimate of 20 million qubits in 8 hours). Moreover, these plans cannot be plain sailing.

在2035年及以后，我们仍然很可能受到量子互联网安全的威胁。然而，不断更新的量子纠错技术仍然有可能解决这一威胁。随着曼哈顿计划制造原子弹，量子技术可能会改变世界。

In 2035 and beyond, we are still very likely to be threatened by the security of quantum Internet. However, the constantly updated quantum error correction technology still has the potential to solve this threat. Quantum technology may change the world as Manhattan plans to build an atomic bomb.

自2006年以来，科学家们一直在开发新一代密码算法，可以抵御量子计算机对现有密码算法的攻击。自2016年以来，NIST对该方法进行了评估。新的密码算法基于量子数字签名（QDs）和密钥封装机制（KEM）。

Since 2006, scientists have been developing a new generation of cryptographic algorithms that can resist the attack of quantum computers on existing cryptographic algorithms. Since 2016, this method has been evaluated by NIST. The new cryptographic algorithm is based on Quantum Digital Signature (QDs) and key encapsulation mechanism (KEM).

通常，我们需要通过链路控制协议执行初始身份验证，交换加密密钥，然后对消息进行加密。今天，我们可以使用RSA 2048+ecdh 256+AES 128来构建公钥加密。在量子互联网时代，我们需要升级到PQC DS+PQC KEM+AES 256加密算法。

Usually, we need to perform initial authentication through the link control protocol, exchange encryption keys, and then encrypt the message. Today, we may use RSA 2048 + ecdh 256 + AES 128 to build public key cryptography. In the era of quantum Internet, we need to upgrade to PQC DS + PQC KEM + AES 256 ENCRYPTION algorithm.

NIST PQC评估——在第一轮中，69个候选算法同时满足最低验收标准和提交要求，其中21个受到损坏或严重攻击。NIST选择26种算法进入第二轮进行更多分析，其中8种算法受到攻击。到2021年，第三轮将完成对七名入围者的评估。该标准草案预计将于2022年发布供公众评论，最终草案将于2024年发布。

NIST PQC evaluation - in the first round, 69 candidate algorithms met both minimum acceptance criteria and submission requirements, of which 21 were damaged or seriously attacked. NIST selected 26 algorithms to enter the second round for more analysis, of which 8 were attacked. By 2021, the evaluation of seven finalists will be completed in the third round. The draft standard is expected to be released in 2022 for public comment, and the final draft will be released in 2024.

现在NIST入围者已经出现，我们可以更清楚地了解中期PQC领域可用的工具。NIST评估过程进展顺利，正在计划中。令人欣慰的是，尽管我们早期担心，但我们相信PQC能够抵御量子计算机的攻击。

Now that NIST finalists have emerged, we can have a clearer understanding of the tools available in the field of medium-term PQC. The NIST assessment process is progressing well and is being planned. It is gratifying that, despite our early worries, we believe that PQC can resist attacks from quantum computers.

随机数是几乎所有密码系统的基本组成部分。QRNG是一种微型量子随机数发生器芯片的原型。Orng已用于一些新兴技术行业商品。IDq及其战略合作伙伴SK Telecom在2020年通过在三星手机上搭载小型QRNG芯片吸引了人们的注意。IDq还与越南智能手机制造商vinsmart签约，并暗示其他手机制造商将很快效仿。国家计划大力支持这些技术的研究和开发。IDq和quside受益于量子技术的旗舰项目，并致力于研究如何使芯片更便携。与此同时，kets还受益于英国的nqtp项目。

Random number is the basic component of almost all cryptosystems. QRNG is a prototype of a subminiature quantum random number generator chip. Orng has been used in some emerging technology industry commodities. IDq and its strategic partner SK Telecom attracted attention by carrying small QRNG chips on Samsung mobile phones in 2020. IDq also signed the Vietnamese smartphone manufacturer vinsmart and hinted that other handset manufacturers would follow suit soon. The state plans to vigorously support the research and development of these technologies. IDq and quside benefited from the flagship project of quantum technology and are committed to studying how to make the chip more portable. At the same time, kets also benefited from the nqtp project in the UK.

量子技术也可以用于在双方之间安全地共享加密密钥。QKD应用于量子力学的基本特性（如量子不可克隆、量子不确定性等），以确保传输过程中任何窃取密钥的企图都会被合法用户发现，这是QKD相对于传统密钥分发的独特优势。

Quantum technology can also be used to securely share encryption keys between two parties. QKD is applied to the basic characteristics of quantum mechanics (such as quantum non cloning, quantum uncertainty, etc.) to ensure that any attempt to steal the key in transmission will be found by legitimate users, which is the unique advantage of QKD over traditional key distribution.

量子密钥分发的安全性是基于量子力学的基本原理，而传统密码学是基于某些数学算法的计算复杂性。传统的密码学无法检测到窃听，无法保证密钥的安全性。QKD只能在传输过程中受到攻击，因此可以用来提供独特的持久安全性。它的主要缺点是需要额外的硬件和成本。在早期不成熟的量子设备中，密钥率不高，必须限制在70-90km的范围内，才能将可信节点用于中继协议。

The advantages and disadvantages of QKD - the security of quantum key distribution is based on the basic principles of quantum mechanics, while traditional cryptography is based on the computational complexity of some mathematical algorithms. Traditional cryptography can't detect eavesdropping, so it can't guarantee the security of key. QKD can only be attacked during transmission, so it can be used to provide unique persistent security. Its main disadvantages are the need for additional hardware and cost. In the early immature quantum devices, the key rate is not high, and the range must be limited to 70-90km, so that the trusted node can be used for relay protocol.

现在，越来越多的公司以盈利模式提供QKD系统。国盾量子为全球最大的运营网络提供硬件服务。目前，京沪干线正在扩建2000公里，扩建5500公里。合肥至武汉的水平主干线已建成700公里，在建项目360公里，拟建项目2200公里。

Now, more and more companies provide QKD system in a profit-making mode. GuoDun quantum provides hardware services for the world's largest operating network so far. At present, 2000 km of Beijing Shanghai trunk line is being expanded and 5500 km of extension line is being built. The construction of 700 km of horizontal trunk line between Hefei and Wuhan has been completed, in addition to 360 km of projects under construction and 2200 km of proposed projects.

在欧洲，euroqci计划包括25个欧盟国家、欧盟委员会和欧空局。其目标是建立一个泛欧洲安全量子通信基础设施。Openqkd在欧洲建立了14个测试中心。这不仅包括电信骨干网和云数据中心等核心领域，还包括智能电网、电子医疗和电子政务等应用领域。

In Europe, the euroqci program includes 25 EU countries, the European Commission and ESA. Its goal is to establish a pan European secure quantum communication infrastructure. Openqkd has established 14 test centers in Europe. These include not only core areas such as telecom backbone network and cloud data center, but also application fields such as smart grid, e-health care and e-government.

在过去两年中，quantum xchange推出了产品PhIO QK（多点QKD）和PhIO TX（提供带外和PQC密钥的嵌入式解决方案），为纽约的金融、电信和政府部门提供服务。

Over the past two years, quantum xchange has launched products PhIO QK (multipoint QKD) and PhIO TX (an embedded solution that provides out of band and PQC keys) to serve the financial, telecommunications and government departments in New York.

要实现真正的量子互联网还有很长的路要走。然而，与当前互联网的出现类似，最初的标准和架构将对未来产生长期影响。现在，了解量子互联网可能的应用领域有助于我们做出更好的决策。

There is still a long way to go to realize the real quantum Internet. However, similar to the emergence of the current Internet, the initial standards and architecture will have a long-term impact on the future. Now understanding the possible application fields of quantum Internet may help us make better decisions.

盲量子计算-一种远程量子计算模式，有望简化客户端所需量子态的传输。在量子时代，"盲"意味着量子服务提供商无法获得计算任务的所有信息，从而确保计算的安全性。

Blind quantum computing - a remote quantum computing mode, which is expected to simplify the transmission of the required quantum states from the client. In the quantum era, "blind" means that quantum service providers cannot obtain all the information of computing tasks, so as to ensure the security of computing.

手持设备-英国的quantum comms hub一直致力于开发便携式QKD设备。其目的是通过创建用于pin保护和身份验证的手持消费设备来增强现有应用程序的安全性，如与ATM的交互。这项技术使量子互联网能够进入未来的智能设备，如非接触式支付、访问控制和数字签名。

Handheld devices - quantum comms hub in the UK has been committed to developing portable QKD devices. The purpose is to enhance the security of existing applications, such as interaction with ATM, by creating handheld consumer devices for pin protection and authentication. This is a technology that enables quantum Internet to enter future intelligent devices, such as contactless payment, access control and digital signature.

6、 量子计时、成像和传感的展望

6、 Prospect of quantum timing, imaging and sensing

通用基础技术在量子技术的各个领域正得到越来越广泛的应用。新产品陆续上市,产品层出不穷。创新领导人正在为量子领域的研究人员树立榜样，证明雄心壮志与收获成正比。量子具有巨大的长期潜力。

General basic technology is developing more and more extensive applications for various fields of quantum technology. New products are on the market one after another, and products will emerge one after another. Innovative leaders are setting an example for researchers in the quantum field to prove that ambition is proportional to harvest. Quantum has great long-term potential.

鉴于量子的广泛潜在应用，量子技术不仅涉及计算和网络，而且在计时、传感和成像方面也发挥着重要作用。研发的挑战通常是将技术带出实验室，并使其在现实生活中实用。

In view of the wide range of potential quantum applications, quantum technology not only involves computing and networks, but also plays an important role in timing, sensing and imaging. The challenge of R & D is usually to bring technologies out of the laboratory and make them practical in real life.

OPMS（光泵磁强计）：squid技术已广泛应用于MEG扫描仪，用于对人脑活动进行无创性研究。基于opmmeg系统的OPMS使传感器不再笨重，可以应用于所有组。

OPMS (optical pump magnetometer): squid technology has been widely used in MEG scanners for non-invasive study of human brain activities. OPMS based on opmmeg system makes the sensor no longer bulky and can be applied to all groups.

经过四年的发展，新成立的Cerca已经使用opm meg扫描仪进行临床评估，以发现更多的临床诊断标志物和新的治疗机会。

After four years of development, start-up Cerca has used opm-meg scanner for clinical evaluation in order to find more clinical diagnostic markers and new treatment opportunities.

一些早期的量子初创公司多年来一直在开发OPM技术。Cerca面临的主要挑战是如何确保头盔的"私人定制"，使受试者戴上头盔后仍然可以自由移动头部，并确保收集数据的准确性。解决这些问题需要大量的技术人才和多样化的技能。国家量子技术计划不仅将为研发资金提供援助，还将建立适当的网络。量子技术投资者需要选择合适的合作伙伴，制定合理的研发计划。

Some early quantum startups have developed OPM technology for many years. The main challenge faced by Cerca is how to ensure the "private customization" of the helmet, so that the subjects can still move their heads freely after wearing the helmet, and ensure the accuracy of the collected data. Solving such problems requires a large number of technical talents and diversified skills. The national quantum technology program will not only provide assistance for R & D funds, but also establish appropriate networks. Quantum technology investors need to select appropriate partners and formulate reasonable R & D plans.

钻石NV色心可在环境温度下工作。虽然灵敏度不高，但它可以小型化，而且它的毒性特性使它特别适合野外生物测量。Qnami受益于量子技术旗舰项目asteriqs的参与。该项目的合作伙伴还包括泰利斯集团、博世、nvision和比利时微电子研究中心。它们各自在钻石技术方面寻求不同的应用。这项技术有望带来许多令人惊讶的用途。

Diamond NV color center can be operated at ambient temperature. Although the sensitivity is not high, it can be miniaturized, and its toxic properties make it especially suitable for field biological measurement. Qnami benefited from the participation of quantum technology flagship project asteriqs. The partners of the project also include Thales group, Bosch, nvision and Belgium Microelectronics Research Center. They each seek different applications in diamond technology. This technology is expected to bring many surprising uses.

Hp mri是一种先进的mri诊断技术，它可以跟踪注入人体的糖并显示糖的变化。例如，这有助于区分报告胸痛患者的活体/无生命心脏组织。然而，由于该方法生产过程中超极化分子的消耗缓慢且昂贵，该技术尚未得到广泛应用。使用diamond NV色心承诺提供更快、更具成本效益和可部署的解决方案。

Hp-mri is an advanced MRI diagnostic technology, which can track the sugar injected into the human body and show what the sugar becomes. For example, this is useful in distinguishing living / inanimate heart tissue in patients reporting chest pain. However, due to the slow and expensive consumption of hyperpolarized molecules in the production of this method, this technology has not been widely used. Using diamond NV color centers promises faster, more cost-effective and deployable solutions.

Metaboliqs是量子技术的旗舰项目，正在寻求开发基于NV diamond的hp mri技术。他们最近从概念验证转向性能提高1000倍的原型。

Metaboliqs, the flagship project of quantum technology, is seeking to develop hp-mri technology based on NV diamond. They recently moved from proof of concept to a prototype with 1000 times better performance.

政府计划在加速技术适应各种应用方面发挥重要作用。此外，投资者不应对分布式量子计算和量子互联网技术中的钻石NV色心感到惊讶。

The government plans to play an important role in accelerating the adaptation of the technology to various applications. In addition, investors should not be surprised by the diamond NV color center in distributed quantum computing and Quantum Internet technology.

志东认为，量子技术可能是21世纪人类技术最重要的突破之一，近年来发展迅速。然而，目前许多量子技术仍处于非常初级的阶段。我们还需要对新技术保持理性态度，不应过度解读和炒作。

Zhidong believes that quantum technology is perhaps one of the most important breakthroughs in human technology in the 21st century, which has developed rapidly in recent years. However, at present, many quantum technologies are still in a very primary stage. We also need to maintain a rational attitude towards new technologies and should not interpret and hype excessively.

Link：https://new.qq.com/omn/20210731/20210731A0BZ8J00.html

update time：2021-08-01 10:37:05