Self-compensating GigaHertz-clocked Quantum Key Distribution

我们所有人的生活都或多或少地被密码学所影响。从童年的简单秘密日记，到网上购物和银行业务，再到政府的国家安全，信息的混乱以隐藏其内容而不被潜在的窃听者窃听是现代生活的一部分。目前最常用的加密形式是所谓的公钥系统，用于电子商务和许多其他应用程序。这种加密方法依赖于将某些数学函数从答案反转到初始初始值所涉及的难度。这类似于烘焙蛋糕，混合原料并制作蛋糕相对容易，但实际上不可能将完成的蛋糕带回最初的原料。然而，并不能长期保证这些数学函数仍然难以逆转。量子计算机的发展将使解密公钥加密信息变得相对容易。幸运的是，可能标志公钥加密系统结束的相同物理也可能以量子密钥分发的形式提供解决方案。量子密钥分发（QKD）利用量子力学提供了一种分发加密或解密消息（密钥）所需信息的方法，这种方法提供了可验证的安全性。它使我们注意到光子（光粒子）的某些性质不能绝对确定。如果利用这些特性将单个二进制数字（位）的信息编码到单个光子上，那么任何窃听密钥交换的窃听者都会干扰对光子的分析，从而在传输过程中留下虚拟指纹，这样发送方和接收方就可以检测到窃听者的存在。QKD系统既可以使用光纤将单光子传输给Bob，也可以通过空气传输。在全球范围内，用于传输电话的光纤基础设施已经进行了大规模投资，因此，QKD系统与该网络兼容可能是至关重要的。然而，设计用于该光纤网络的最佳传输波长的探测器尚未达到其他探测器可用的性能水平，导致加密密钥交换的最大速率大幅降低。我们已经能够使用设计用于更短波长的更先进的探测器来开发一个现有的系统，用于校园规模的实施，与电信光纤完全兼容。该QKD系统能够传输比特，每秒产生数百万比特，据我们所知，这是目前存在的最快时钟速率QKD系统。在这些比特传输速率下，经过传输和纠错技术，实时加密视频会议成为可能。该系统还被独特地用于多用户（一个Alice和多个bob）使用。尽管在世界级的密钥汇率下运行，但该系统确实存在长期稳定性方面的严重问题，并且基于一个可能不太安全的QKD协议。我们建议利用在现有系统开发中获得的专业知识构建一系列新的，可能更稳定的QKD系统，所有这些系统都将使用更安全的协议。前两个系统将基于现有的设计，已经在更长的波长下工作，但密钥交换率（即每秒几千比特）要低得多。第三个系统将基于一种新颖的设计，这种设计从未完全实现，但在防范窃听攻击方面可能具有更高的安全性。对于这三个系统，将调查多用户操作的可能性，并将进行多用户网络的示范。

All of our lives have been touched by cryptography at some point or another. From the simple secret diaries of childhood, through internet shopping and banking to the national security of governments, the scrambling of information to hide its contents from potential eavesdroppers is part of modern life.The most commonly used form of encryption today is the so-called public key system, as used in e-commerce and many other applications. This encryption method relies on the difficulty involved in reversing certain mathematical functions back from an answer to the initial starting values. This is analogous to baking a cake, where it is relatively easy to mix the ingredients and produce a cake but virtually impossible to take a finished cake and return to the initial ingredients. However, there is no long-term guarantee that these mathematical functions will remain as difficult to reverse. The development of quantum computers would allow relatively easy decryption of public key encrypted messages. It is fortunate that the same physics that may signal the end of public key encryption systems may also provide the solution in the form of quantum key distribution.Quantum key distribution (QKD) uses the science of quantum mechanics to provide a means of distributing the information needed to encrypt or decrypt a message (the key) in a way that provides verifiable security. It makes us of the fact that certain properties of a photon (a light particle ) cannot be determined with absolute certainty. If single binary digits (bits) of information are encoded onto single-photons using these properties, then any eavesdropper listening in on the key exchange will disturb the analysis of photons sufficiently to leave a virtual fingerprint on the transmission so that sender and receiver can detect the presence of the eavesdropper.QKD systems can either operate using optical fibres to transmit the single-photons to Bob or transmit them through the air. There has already been a large scale investment in a worldwide infrastructure of optical fibre to transmit telephone calls, so it could prove critical that a QKD system would have compatibility with this network. However, the detectors designed for use at the optimum transmission wavelengths for this fibre network have not yet attained the level of performance available from other detectors, leading to a massive reduction in the maximum rate of cryptographic key exchange.We have been able to use more advanced detectors designed for use at shorter wavelengths to develop an existing system for use in a campus sized implementation that is fully compatible with the telecommunications optical fibre. This QKD system is capable of transmission of bits generating up to several million bits per second, which is, to the best of our knowledge, the fastest clock rate QKD system currently in existence. At these bit transmission rates, after transmission and error correction techniques, real-time encrypted video conferencing becomes a possibility. This system has also been uniquely adapted for multi-user (one Alice and multiple Bobs) use.Although operating at world-class key exchange rates, this system does suffer from serious issues with long-term stability and is based on an arguably less secure QKD protocol. We propose to construct a series of three new, potentially much more stable, QKD systems using the expertise gained in the development of our existing system, all of which will use an arguably more secure protocol. The first two systems will be based on existing designs already operating at longer wavelengths but at much lower key exchange rates (ie a few thousands of bits per second). The third system will be based around a novel design that has never been fully implemented but has potentially greater security from eavesdropping attacks. With all three systems, the potential for multi-user operation will be investigated and demonstrations of multi-user networks will be made.