Improved fluxonium readout through dynamic flux pulsing

by Taryn V. Stefanski, Figen Yilmaz, Eugene Y. Huang, Martijn F.S. Zwanenburg, Siddharth Singh, Siyu Wang, Lukas J. Splitthoff, and Christian Kraglund Andersen

Data availability

To be made publicly available.

Related work

Stefanski, T. V., & Andersen, C. K. (2024). Flux-pulse-assisted readout of a fluxonium qubit. Physical Review Applied, 22(1), 014079.

Requirements

• [Python == 3.8.8]
• [Cython == 0.29.20]
• [NumPy == 1.23.4]
• [SciPy == 1.8.1]
• [QuTiP == 4.7.0]
• [scQubits == 3.1.0]
• [Quantify-core == 0.7.4]
• [Matplotlib == 3.7.3]
• [Pandas == 1.2.4]

Description

Relevant code used for data analysis and numerical modeling of flux-pulse-assisted readout. Brief explanations of files given below.

.csv files:

• res_fit_results.csv : contains fitted quantities from readout resonator spectroscopy with qubit prepared in |0> or |1>. The data was fit using the resonator magnitude model provided by our proprietary library from Orange Quantum Systems. This .csv file contains the resonator frequencies, linewidths, and dispersive shifts as a function of DC current bias converted to external flux bias in units of magnetic flux quanta, and the resonator frequency when the qubit is prepared in |0> as a function of flux pulse amplitude. Relevant data folders: 20231110, 20231111, 20231112, 20231113.
• processed_data_fx8.csv : exported quantities produced by process_fluxonium_data.ipynb.
• RO.csv : Measured data presented in Fig. 3 showing assignment and SNR-limited fidelity as a function of integration time at sweet-spot and using flux-pulse-assisted readout with 400mV readout amplitude, equivalent to ~52 resonator photons.

.ipynb files:

• process_fluxonium_data.ipynb : imports fitted resonator data from res_fit_results.csv, extracts measured qubit frequency and relaxation time as a function of external flux bias, and uses scQubits to model a fluxonium coupled to a resonator in order to fit the data and extract Ej, Ec, El, g, and (bare) resonator frequency. These values, along with qubit frequency and T1, are exported to processed_data_fx8.csv. Relevant data folders: 20231110, 20231111, 20231112, 20231113.
• exp_dispersive_shift_fitted.ipynb : imports res_fit_results.csv and processed_data_fx8.csv. Extracted energy parameters used to model dispersive shift and qubit-resonator detunings using code from https://github.com/AndersenQubitLab/FluxPulseAssistedFluxonium. Resonator frequency, qubit frequency, and dispersive shift data plotted in comparison to model prediction.
• process_FPARO_data.ipynb : imports and plots readout fidelity versus flux pulse amplitude for 5 different readout powers (100, 150, 200, 300, 400 mV), readout fidelity versus readout frequency at sweet-spot, and readout fidelity as a function of integration time for 5 different powers with the same, and variable, flux pulse amplitudes. Data pertaining to readout as a function of integration time using the highest readout amplitude (400 mV) which is shown in Fig. 3 is exported to RO.csv. Relevant data folders: 20231115, 20231116, 20231117, 20231118, 20231119, 20231201.
• readout_dynamics_numerics.ipynb : imports data from processed_data_fx8.csv, res_fit_results.csv, and RO.csv to compare measured data with theory predictions from numerically solving the Langevin equation to obtain the intracavity field and resulting output field, adapted from the work in 'Flux-pulse-assisted readout of a fluxonium qubit' (reference above). This generates the numerical curves of Fig. 3.
• meas_efficiency.ipynb : imports, plots, and analyzes coherence data, i.e. variable strength single-shot readout and Ramsey versus measurement pulse amplitude. Subsequently extracts measurement efficiency and imports data from res_fit_results.csv to convert readout amplitude to average resonator photon number. Additionally imports res_fit_results.csv to provide conversion of flux pulse amplitude to effective external flux bias in units of magnetic flux quanta. Relevant data folders: 20231129

.py files

• ROfunctions.py : contains functions from https://github.com/AndersenQubitLab/FluxPulseAssistedFluxonium for diagonalizing full Hamiltonian, truncating, and calculating dispersive shift and detunings. Used in exp_dispersive_shift_fitted.ipynb.
• SSRO_functions.py : contains functions for processing single shot readout data. Used in process_FPARO_data.ipynb and meas_efficiency.ipynb.