Noise resilient quantum computing: GKP codes and protected superconducting qubits
Recent experiments have demonstrated quantum computing below the threshold for quantum error correction, meaning that algorithmically relevant error rates can be achieved by scaling the quantum computer. However, reaching these error rates with current hardware will require a very large overhead in the number of physical qubits. To reduce this overhead, the noise resilience of the physical qubits must be improved. In this talk, I will discuss approaches to improving this noise resilience at the hardware level for the three main steps of quantum computing: state preparation, gates and measurement.
In the first part of this talk I will discuss how bosonic codes states, namely GKP states, can be prepared deterministically in a superconducting circuit without an ancillary qubit. This represents an alternative to standard approaches, which are limited by errors that propagate from the noisy ancillary qubit. Here we utilize Floquet engineering to simplify the hardware required to realize the GKP Hamiltonian. Adiabatic state preparation can then be used to prepare these states from the ground states of a harmonic oscillator. Encoding qubits into bosonic codes such as the GKP code offers a layer of noise protection prior to concatenating to a larger DV code. As such, our proposal offers a hardware-efficient approach to reducing physical error rates during state preparation.
In the next part of the talk, I will discuss a way to perform a protected gate with a protected superconducting qubit, namely the 0-π qubit. Similarly to bosonic encodings, these qubits offer lower physical error rates by reducing the qubit’s sensitivity to noise. However, this reduced sensitivity comes at the cost of decreased controllability, as approaches for performing gates with standard superconducting qubits are ineffective with these noise-protected qubits. To overcome this whilst maintaining the qubit’s protection, an ancillary bosonic mode can be used to perform a fault-tolerant gate through a GKP encoding. Crucially, our scheme utilizes an internal mode of the 0-π qubit as the ancillary bosonic mode, yielding a parsimonious approach to controlling these qubits.
Finally, I will discuss QND measurement schemes for protected qubits in two conjugate bases and give a concrete implementation for the 0-π qubit. We again utilize the internal bosonic mode of the 0-π qubit, as well as conditional displacement gates to measure encoded GKP states. Together with the protected phase gate, these measurements provide a route for universal fault-tolerance with protected qubits.
The first two sections of this talk are based on the published articles Phys. Rev. Lett. 132, 130605 (2024) [arXiv:2303.03541] and PRX Quantum 7, 010306 (2026) [arXiv:2503.14634], whilst the final part of the talk discusses ongoing work.