Computer Engineering and Applications

• Theory, Research and Development • Previous Articles     Next Articles

Comprehensive SWAP Optimization Strategy for Improving Initial Qubit Mapping

LI Hui, HAN Zi’ao, LU Kai, LIU Shujuan, JU Mingmei   

  1. 1.School of Computer and Information Engineering, Harbin University of Commerce, Harbin 150028, China
    2.Heilongjiang Key Laboratory of Electronic Commerce and Information Processing, Harbin 150028, China
  • Online:2024-07-15

改进量子位初始映射的综合SWAP优化策略

李晖,韩子傲,卢凯,刘述娟,鞠明媚   

  1. 1.哈尔滨商业大学 计算机与信息工程学院,哈尔滨 150028
    2.黑龙江省电子商务与信息处理重点实验室,哈尔滨 150028

Abstract: In the noisy intermediate-scale quantum (NISQ) era, traditional initial mapping strategies overlook potential adjacency issues in subsequent operations. To address this challenge, a multi-factor interaction cost function is devised, considering the distance between qubits, the interaction time, and the gate operations error rate. Additionally, the comprehensive SWAP optimization strategy (CSOS) is proposed to address the adjacency problem in quantum circuits, including the optimal SWAP selection and the SWAP-based batch update strategy (BUS) to optimize the local qubit mapping in quantum circuits. Optimal SWAP selection involves SWAP operations choices with the highest benefit by comparing advantages. The BUS considers upcoming quantum operation sequences during the mapping phase and proactively performs SWAP operations batches. These approaches reduce the SWAP operation numbers during the entire circuit execution process, minimizing mapping overhead to the greatest extent. Experimental results demonstrate that the CSOS reduce the average number of inserted SWAP gates by 38.1% and lower hardware gate count expenses by approximately 12%.

Key words: quantum computing, initial mapping, comprehensive SWAP optimization strategy (CSOS), best SWAP choice, batch updating

摘要: 在嘈杂的中尺度量子时代,传统的初始映射策略忽略了后续操作中可能出现的邻接性的问题。针对这一挑战,综合考虑量子比特间的距离、交互时间和门操作的错误率,设计了一个多因素交互成本函数,并提出综合SWAP优化策略(comprehensive SWAP optimization strategy,CSOS)。该策略包括最佳SWAP选择和基于SWAP的批量更新策略,用于优化量子电路的局部量子位映射。最佳SWAP选择通过对比SWAP操作的效益,选择最佳收益的SWAP门;批量更新策略在映射阶段考虑即将执行的量子操作序列,预先执行批量的SWAP操作。二者综合可以减少整个电路执行过程中的SWAP数量,以最大程度减少映射开销。实验结果显示,CSOS优化方式可以平均减少38.1%的插入SWAP门数量,并降低约12%的硬件门计数开销。

关键词: 量子计算, 初始映射, 综合SWAP优化策略, 最佳SWAP选择, 批量更新