Integrated Optics as a Platform for Continuous Variable Measurement Based Quantum Computing

最近已经有使用铌酸锂和硅基集成芯片架构的挤压的演示[5]。这些结果被认为对量子计算是有用的，但是到目前为止所达到的压缩水平很低。最近，极大的2D连续变量团簇态已经使用体光学产生[1，2]。这些演示的一个关键要素是大的时间延迟，以支持使用时间箱编码。增加集群中输入模式的数量需要增加延迟，从而限制了输入寄存器的可扩展性。该博士旨在探索使用集成光学技术生成CV簇状态，以期设计可扩展的集成平台。要解决的关键实验挑战是簇状态验证（~3-5dB）和完全容错（~15- 17 dB [7]）所需的压缩水平与当前集成结果的调和。因此，博士学位的第一阶段将是使用商业上可用的锂离子电池模块耦合到具有集成零差检测器的芯片中，以观察所需的最小3dB的挤压。博士是足够灵活的枢轴和使用证明硅和氮化硅为基础的挤压，如果他们满足要求。一旦这已经实现，博士的一个里程碑将是理解的理论，设计，构建和验证基于现有的散装方案的1D集成大规模集群状态生成平台[3]。这个项目的一个延伸目标是在集群状态上实现高斯运算，以证明最近建立的技术[4]的计算概念证明。第二个目标项目是将集群状态扩展到2D。然而，由于双时间仓编码方法中所需的大延迟（和相关损失），将需要探索使用频率和/或空间模式的替代平台[6]。博士学位的这个子主题将构成项目的多种可能性，包括在两个光谱分离的频率模式之间进行状态隐形传态，并使用生成的实验进行多方纠缠演示。最后，博士学位的另一个方向是研究CV计算中容错所需的非高斯状态的光学生成。除了实验项目外，博士学位还将包括与通用CV-MBQC容错相关的重要理论组成部分。为了利用纠错技术，需要玻色子码来将量子比特嵌入连续系统中。此外，为了实现普适性，需要非高斯运算。这两个挑战都可以通过使用注入到簇状态中的“GKP”量子位来解决[8]。与Nicolas Menicucci合作，通过研究混合高斯/ GKP编码方案直接解决其中一些挑战。拟议的合作将涉及借调到RMIT在尼克的小组内3-6个月来开发这个项目。二维聚类状态的确定性生成https://science.sciencemag.org/content/366/6463/3692.生成时域复用的二维簇状态。https://science.sciencemag.org/content/366/6463/3733通过无限时域复用生成百万模式连续变量集群状态https：//aip.scitation.org/doi/10.1063/1.4962732 [4]. 4.基于一百步测量的量子计算在时域中以25 MHz时钟频率复用https://arxiv.org/pdf/2006.11537.pdf5。纳米光子器件中使用双泵浦四波混频的单模正交压缩。https://arxiv.org/abs/2001.094746 https://arxiv.org/pdf/1912.112157. https://journals.aps.org/pra/abstract/10.1103/PhysRevA.100.0103018. https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.123.200502

Recently there have been demonstrations of squeezing using Lithium Niobate and Silicon-based integrated chip architectures[5]. These results have been claimed to be useful for quantum computing, however the squeezing levels achieved so far are low. Also recently, extremely large 2D continuous variable cluster states have been generated using bulk optics[1,2]. A key element of these demonstrations were large temporal delays to enable the use of time-bin encoding. Increasing the number of input modes in the cluster requires an increase in the delay, constraining the scalability of an input register. This PhD aims to explore the generation of CV cluster states using integrated optical technology with the view to designing a scalable integrated platform.A key experimental challenge to be addressed is the reconciliation of the squeezing levels required for cluster state verification (~3-5dB) and full fault tolerance (~15-17dB [7]) with current integrated results of ~1-2dB. A first stage of the PhD would therefore be to use commercially available lithium niobite modules coupled into a chip with integrated Homodyne detectors to observe the minimum 3dB of squeezing required. The PhD is sufficiently flexible to pivot and use demonstrated silicon and silicon-nitride based squeezing if they meet requirements.Once this has been achieved, a milestone of the PhD will be to understand the theory of, design, construct and verify a 1D integrated massive cluster state generation platform based upon existing bulk schemes[3]. A stretch goal of this project would then be to implement gaussian operations on the cluster state to demonstrate a computing proof of concept following recently established techniques[4]. A second targeted project will be to extend the cluster state to 2D. However, due to the large delays required (and associated losses) in the double time-bin encoding method an alternative platform using frequency and/or spatial modes will need to be explored[6]. This sub theme of the PhD will constitute multiple possibilities for projects, including performing state teleportation between two spectrally separate frequency modes and using the generated experiments for multi-partite entanglement demonstrations. Finally, an alternate direction for the PhD would be to look at optical generation of required non gaussian states for fault tolerance in CV computation.In addition to the experimental projects, the PhD will also consist of a significant theory component relating to Fault Tolerance for Universal CV-MBQC. To take advantage of error correction techniques, a bosonic code is needed to embed a qubit within the continuous system. Furthermore, to enable universality a Non-Gaussian operation is needed. Both of these challenges can be resolved by the use of 'GKP' qubits which are injected into the cluster states[8]. A collaborative effort with Nicolas Menicucci is proposed to directly address some of these challenges by studying mixed Gaussian / GKP encoding schemes. The proposed collaboration will involve a secondment at RMIT within Nick's group for 3-6 months to develop this project.1. Deterministic generation of a two-dimensional cluster state https://science.sciencemag.org/content/366/6463/3692. Generation of time-domain-multiplexed two-dimensional cluster state https://science.sciencemag.org/content/366/6463/3733. Generation of one-million- mode continuous-variable cluster state by unlimited time-domain multiplexing https://aip.scitation.org/doi/10.1063/1.4962732[4]. 4. One-hundred step measurement-based quantum computation multiplexed in the time domain with 25 MHz clock frequency https://arxiv.org/pdf/2006.11537.pdf5. Single-mode quadrature squeezing using dual-pump four-wave mixing in a nanophotonic device https://arxiv.org/abs/2001.094746. https://arxiv.org/pdf/1912.112157. https://journals.aps.org/pra/abstract/10.1103/PhysRevA.100.0103018. https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.123.200502