Chalmers team uses Quantum Lattice Gates to run bosonic operations up to 1,000× faster

Bosonic quantum codes store qubits in microwave resonator fields rather than individual transmons. Those encoded states offer built-in protection against some errors, but preparing and controlling them has relied on slow adiabatic ramps. Longer operations expose fragile states to environmental noise and degrade error correction performance. Completing operations in a single Floquet period reduces exposure time and the cumulative effect of noise.

• Binomial and 4-component cat code preparations in the same fidelity regime.
• Universal single-qubit logical gates (Hadamard, Phase, π/8) with average gate errors on the order of 10⁻³ executed within microsecond time windows.
• Noise robustness roughly three orders of magnitude better than comparable adiabatic-ramp (AR) implementations.
• Scaling of resource demand linear in Hilbert-space dimension O(D), including efficient Haar-random state sampling.

The protocol was designed with superconducting resonator platforms in mind. It is compatible with existing superconducting circuits and with the hardware goals stated for the Wallenberg Centre for Quantum Technology (WACQT), which is building a 100-qubit superconducting processor. The method's linear scaling with Hilbert-space dimension suggests a practical path to larger bosonic encodings without exponential resource growth.

Implications for quantum error correction and next steps

Faster, single-period operations reduce the time that encoded information is vulnerable during state preparation and gate execution. That directly benefits quantum error correction (QEC) by lowering the window for decoherence and correlated noise to act. The study reports that combining QLGs with OPE yields high-fidelity preparations and gates suitable for QEC routines. The work is presented as a theoretical and computational advance published in Physical Review Letters.

What remains to be addressed experimentally

The results reported are theoretical and computational demonstrations targeted at superconducting resonator hardware. Experimental validation on physical devices will be needed to confirm achievable fidelities and robustness under real-device noise and control imperfections. The next practical steps will be implementing these driven Floquet cycles and QLG controls in laboratory superconducting systems and integrating them into QEC experiments.