Linear depth approximate QFT with zero ancilla

The inseparable coupling of a quantum system with its environment adds noise at resonant frequencies, causing energy fluctuations and losses entanglement of the fragile quantum state, informally known as decoherence. By truncating Fourier wiggles with the fault-tolerant features, a robust quantum Fourier transform (QFT) scheme can mitigate the decoherence effect. However, a low qubit-count and circuit-depth-based quantum circuit is desirable because of decoherence rate scale up linearly with qubit-count, shortening the coherence time of a fragile quantum state. Indeed, due to the high latency of 1-qubit phase-gates, phase-depth becomes a more dominant factor in circuit-depth and pays a high price for scalability. The above issues are addressed in this manuscript by implementing a fault-tolerant approximate quantum Fourier-transform (AQFT) in linear phase-depth with zero ancillary cost and \(O(nlog n)\) phase-count using the Clifford+ \(Z_N\) -group. By truncating lower rotated phase operators, a robust fault-tolerant AQFT is achieved with linear phase-depth and zero ancillary cost, as well as 75%, 87.6%, 94.22%, and 26.99% average improvements in phase-depth, phase-count, CNOT-count, and ancillary cost respectively over the present-state-of-the-arts 1 . The proposed AQFT maintains polynomially small approximation error, achieving a strong balance between accuracy and efficiency that makes it practical for most algorithms and well suited for NISQ hardware.