Establish fidelity scaling for large-dimensional native qudit–oscillator interactions

Establish the fidelity scaling of the native trapped-ion qudit–oscillator interactions as the qudit dimension increases, in order to determine whether the architecture remains experimentally viable for large multilevel registers.

Background

The proposed architecture relies on spectrally resolved multilevel control within a trapped ion, with larger experimentally usable dimension allowing larger discrete manifolds to be encoded in a single ion. However, increasing the number of levels also increases the requirements for spectral addressing, calibration, and leakage suppression. The paper validates the instruction set on a small ququart benchmark but does not experimentally establish how the fidelity of its native qudit–oscillator interactions changes at larger qudit dimensions. Resolving this issue is necessary to assess the practical scaling advantage of the architecture beyond the demonstrated regime.