Optimized Gottesman-Kitaev-Preskill Error Correction via Tunable Preprocessing

Abstract

The Gottesman-Kitaev-Preskill (GKP) code is a promising bosonic candidate for realizing fault-tolerant quantum computation. Among existing error-correction protocols for GKP code, the Steane-type scheme is a canonical and widely adopted paradigm, yet its intrinsic noise propagation pattern limits further performance improvement. In this work, we propose a preprocessing-based Steane-type (P-Steane) scheme, which introduces a tunable preprocessing stage with squeezing parameters $a$ and $b$ to actively reshape noise propagation, thereby constituting a parameter framework. This framework spans a spectrum of protocols beyond existing methods, reproducing the performance of both the ME-Steane scheme ($a=1$, $b=1$) and the teleportation-based scheme ($a=1/\sqrt{2}$, $b=\sqrt{2}$) as special cases. Crucially, in the small-noise regime and when the data qubit is noisier than the ancilla qubits, P-Steane scheme achieves the minimum product of position- and momentum-quadrature output noise variances when $2a = b$, and consistently outperforms the ME-Steane scheme within a specific squeezing-parameter range under this condition.