It has previously been established that the logarithmic-depth approximate quantum Fourier transform (AQFT) provides a suitable replacement for the regular QFT in many quantum algorithms. Since the AQFT is less accurate by definition, polynomially many more applications of the AQFT are required to achieve the original accuracy. However, in many quantum algorithms, the smaller size of the AQFT circuit yields a net improvement over using the QFT. This paper presents a more thorough analysis of the AQFT circuit, which leads to the surprising conclusion that for sufficiently large input sizes, the difference between the QFT and the logarithmic-depth AQFT is negligible. In effect, the AQFT can be used as an direct replacement for the QFT, yielding improvements in any application which does not require exact quantum computation.
We investigate resource requirements for a fault-tolerant quantum Fourier transform. The quantum Fourier transform is a basic subroutine for quantum algorithms which provide an exponential speedup over known classical ones, such as Shor's algorithm for factoring. To implement single-qubit rotations required for a quantum Fourier transform in a fault-tolerant manner, we consider two types of approaches: gate synthesis and state distillation. While the gate synthesis approximates single-qubit rotations with basic quantum operations, the state distillation allows one to perform single-qubit rotations for a quantum Fourier transform exactly. It is unknown, however, which approach is better for a quantum Fourier transform. Here we develop a state-distillation method optimized for a quantum Fourier transform and compare this performance with those of state-of-the-art techniques for gate synthesis without and with ancillary states (ancillas). The performance is evaluated with the resource requirement for a quantum Fourier transform. The resource is measured by the total number of $\ensuremath{\pi}/8$ gates denoted by
T, which is called the
T count. Contrary to the expectation, the
T count for the state distillation is considerably larger than those for the ancilla-free and ancilla-assisted gate synthesis. Thus, we conclude that the ancilla-assisted gate synthesis is a better approach to a fault-tolerant quantum Fourier transform.