Efficient Preparation of Large Block Code Ancilla States for Fault-tolerant Quantum Computation

TL;DR

This paper presents a fault-tolerant, efficient protocol for preparing large block stabilizer ancilla states crucial for fault-tolerant quantum computation, overcoming correlated error issues with a novel postselection method.

Contribution

It introduces a fault-tolerant ancilla preparation protocol using classical error detection to remove correlated errors, significantly improving yield for large block codes.

Findings

Protocol achieves near-perfect ancilla fidelity with low gate failure rates.

Yield rate improves from O(t^{-2}) to O(1) for large CSS codes.

Numerical simulations validate the protocol's effectiveness for the Golay code.

Abstract

Fault-tolerant quantum computation (FTQC) schemes that use multi-qubit large block codes can potentially reduce the resource overhead to a great extent. A major obstacle is the requirement of a large number of clean ancilla states of different types without correlated errors inside each block. These ancilla states are usually logical stabilizer states of the data code blocks, which are generally difficult to prepare if the code size is large. Previously we have proposed an ancilla distillation protocol for Calderbank-Shor-Steane (CSS) codes by classical error-correcting codes. It was assumed that the quantum gates in the distillation circuit were perfect; however, in reality, noisy quantum gates may introduce correlated errors that are not treatable by the protocol. In this paper, we show that additional postselection by another classical error-detecting code can be applied to remove almost all correlated errors. Consequently, the revised protocol is fully fault-tolerant and capable of preparing a large set of stabilizer states sufficient for FTQC using large block codes. At the same time, the yield rate can be boosted from $O(t^{-2})$ to $O(1)$ in practice for an $[[n,k,d=2t+1]]$ CSS code. Ancilla preparation for the $[[23,1,7]]$ quantum Golay code is numerically studied in detail through Monte Carlo simulation. The results support the validity of the protocol when the gate failure rate is reasonably low. To the best of our knowledge, this approach is the first attempt to prepare general large block stabilizer states free of correlated errors for FTQC in a fault-tolerant and efficient manner.