Automated Generation of Shuttling Sequences for a Linear Segmented Ion Trap Quantum Computer

TL;DR

This paper presents algorithms for automating the design of shuttling schedules and qubit assignments in segmented ion trap quantum computers, improving efficiency for complex quantum circuits.

Contribution

It introduces a fully automated shuttling schedule generation algorithm and compares different qubit assignment methods for various circuit types.

Findings

Advanced assignment algorithms reduce shuttling overhead for fixed-structure circuits.

Automated scheduling complies with trap constraints.

Effective qubit assignment improves circuit execution efficiency.

Abstract

A promising approach for scaling-up trapped-ion quantum computer platforms is by storing multiple trapped-ion qubit sets ('ion crystals') in segmented microchip traps and to interconnect these via physical movement of the ions ('shuttling'). Already for realizing quantum circuits with moderate complexity, the design of suitable qubit assignments and shuttling schedules require automation. Here, we describe and test algorithms which address exactly these tasks. We describe an algorithm for fully automated generation of shuttling schedules, complying to constraints imposed by a given trap structure. Furthermore, we introduce different methods for initial qubit assignment and compare these for random circuit (of up to 20 qubits) and quantum Fourier transform-like circuits, and generalized Toffoli gates of up to 40 qubits each. We find that for quantum circuits which contain a fixed structure, advanced assignment algorithms can serve to reduce the shuttling overhead.