2023/12/17 by Yuwei Jin, Jin, Yuwei, Xiangyu Gao +11 · 2 citations
Computer Science · Engineering · #FOS: Physical sciences #Low-power high-performance VLSI design #Parallel Computing and Optimization Techniques #Quantum Computing Algorithms and Architecture #Quantum Physics (quant-ph)
paper · pdf · doi:10.48550/arxiv.2312.16114
openalex publication_date 2023/12/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this research paper, our primary focus revolves around the domain-specific hardware mapping strategy tailored for Quantum Fourier Transformation (QFT) circuits. While previous approaches have heavily relied on SAT solvers or heuristic methods to generate hardware-compatible QFT circuits by inserting SWAP gates to realign logical qubits with physical qubits at various stages, they encountered significant challenges. These challenges include extended compilation times due to the expansive search space for SAT solvers and suboptimal outcomes in terms of the number of cycles required to execute all gate operations efficiently. In our study, we adopt a novel approach that combines technical intuition, often referred to as "educated guesses," and sophisticated program synthesis tools. Our objective is to uncover QFT mapping solutions that leverage concepts such as affine loops and modular functions. The groundbreaking outcome of our research is the introduction of the first set of linear-depth transformed QFT circuits designed for Google Sycamore, IBM heavy-hex, and the conventional 2-dimensional (2D) grid configurations, accommodating an arbitrary number of qubits denoted as 'N'. Additionally, we have conducted comprehensive analyses to verify the correctness of these solutions and to develop strategies for handling potential faults within them.