作为标定量子器件整体性能的一种重要方法，量子随机基准（RB）在近未来中小型量子处理器件的性能诊断中将发挥重要作用。然而由于器件集成设计及环境噪声的特性，涉及多比特的操控误差和噪声影响往往存在复杂的关联效应，对RB的直接运用及结果解释造成困难。本项目拟通过对称性度量引入等价性关系对多比特噪声进行系统分类，由此给出各类噪声下RB模型约化的一般性原则。此外，该项目还将以固态自旋量子比特为例阐述新方案在处理电荷噪声问题以及主流可扩展方案中的具体运用。本项目的研究结果对于提高对多比特量子系统的整体表征以及精确操控具有重要意义。

Randomized benchmarking (RB) has been one of the most promising ways for systematical characterization of quantum devices, particularly because of the near-future trends in developing scalable quantum processing devices. Meanwhile, due to the intrinsic features associated with the architectures’ design and the consequent noises, both coherent errors and environmental noises could display sophisticated correlation effects that hinder the application of RB as well as its result interpretation. In this project, a systematic classification of multi-qubits’ noises will be explored based on the equivalence relations that defined with the help of symmetry measures. Equipped with these classifications, a generic method will be pursued for the reduction of RB fitting models involving multi-qubits. Furthermore, application of this reduction approach will be applied to concrete examples appearing in popular solid-state based spin-qubit architectures, with a particular aiming at understanding the charge noise effects and how to reduce them under scalable conditions. Results of this project will benefit the characterization as well as precise control of multi-qubits quantum devices.