High-rate quantum LDPC codes for long-range-connected neutral atom registers

TL;DR

This paper proposes high-rate LDPC quantum error correction codes suitable for neutral atom quantum computers, demonstrating their advantages over surface codes at low error rates and outlining feasible near-term implementations with limited long-range interactions.

Contribution

It introduces a family of high-rate LDPC codes with limited long-range interactions tailored for neutral atom platforms, showing their superior performance in simulations.

Findings

Codes outperform surface codes below 0.1% two-qubit error rate

Feasible implementation using Rydberg blockade in 2D neutral atom arrays

Potential for scalable, low-overhead fault-tolerant quantum computing

Abstract

High-rate quantum error correcting (QEC) codes with moderate overheads in qubit number and control complexity are highly desirable for achieving fault-tolerant quantum computing. Recently, quantum error correction has experienced significant progress both in code development and experimental realizations, with neutral atom qubit architecture rapidly establishing itself as a leading platform in the field. Scalable quantum computing will require processing with QEC codes that have low qubit overhead and large error suppression, and while such codes do exist, they involve a degree of non-locality that has yet to be integrated into experimental platforms. In this work, we analyze a family of high-rate Low-Density Parity-Check (LDPC) codes with limited long-range interactions and outline a near-term implementation in neutral atom registers. By means of circuit-level simulations, we find that these codes outperform surface codes in all respects when the two-qubit nearest neighbour gate error probability is below $\sim 0.1\%$. By using multiple laser colors, we show how these codes can be natively integrated in two-dimensional static neutral atom qubit architectures with open boundaries, where the desired long-range connectivity can be targeted via the Rydberg blockade interaction.