AI 中文总结
本工作分析异质噪声下的分布式格点手术，以XX合并操作为重点，用MWPM解码器表征逻辑错误并估计阈值，推导体与 seam 错误率近似电路级噪声模型，为分布式量子架构的关键参数选择提供实用见解。
AI 中文摘要
分布式量子计算通过共享纠缠对将小型量子模块联网，为单片量子处理器提供了可扩展的替代方案，该场景的核心挑战是模块间量子操作通常比模块内本地门噪声更大，会给系统引入额外噪声。本工作分析了异质噪声条件下的分布式格点手术，重点关注其基本子程序之一的合并操作，具体讨论了两个不同量子处理器上承载的两个旋转表面码块之间的XX合并操作，表征了所得H形时空图中的逻辑错误，并使用最小权完美匹配（MWPM）解码器估计阈值。我们采用唯象噪声模型，推导了不同的体错误率和 seam 错误率，以近似包含本地CNOT门、噪声纠缠对、空闲错误和读出错误贡献的电路级噪声模型。研究结果为分布式架构中选择最优表面码距离、确立目标本地门保真度以及确定逻辑运算所需的可容忍纠缠对保真度提供了实用见解。
英文摘要
Distributed quantum computing offers a scalable alternative to monolithic quantum processors by networking smaller quantum modules through shared entangled pairs. A central challenge in this setting is that inter-module quantum operations are typically noisier than intra-module local gates, which introduces additional noise into the system. In this work, we analyze distributed lattice surgery under heterogeneous noise conditions, focusing in particular on the merge operation as one of its fundamental subroutines. Specifically, we discuss the XX merge operation between two rotated surface-code patches hosted on two different quantum processors. We characterize logical errors in the resulting H-shaped spacetime diagram and estimate thresholds using a minimum-weight perfect matching (MWPM) decoder. We use a phenomenological noise model and derive distinct bulk and seam error rates to approximate a circuit-level noise model that includes contributions from local CNOT gates, noisy entangled pairs, idle errors, and readout errors. Our results provide practical insights into selecting the optimal surface-code distance, establishing target local-gate fidelities, and determining the tolerable entangled-pair fidelity required for logical operations in a distributed architecture.
CommentsAccepted at the IEEE International Conference on Quantum Computing and Engineering (QCE 2026)
