Quantum Noise Limits Fidelity in Trapped Atom Qubit Systems.

Experimental validation confirms the relationship between control noise and qubit state fidelity within a 10×10 optical tweezer array utilising rubidium-85 atoms. Measured fidelities align with theoretical predictions from the stochastic Schrödinger equation, offering insight into noise identification and control optimisation for near-term quantum devices.

The pursuit of scalable quantum computation necessitates a detailed understanding of the imperfections inherent in physical systems. Environmental disturbances and control imprecision introduce noise, which degrades the performance of quantum bits, or qubits, and limits the duration of coherent quantum operations. Researchers at Eindhoven University of Technology, led by D.A. Janse van Rensburg, R.J.P.T. de Keijzer, R.C. Venderbosch, Y. van der Werf, J.J. del Pozo Mellado, R.S. Lous, E.J.D. Vredenbregt, and S.J.J.M.F. Kokkelmans, present experimental validation of a theoretical framework linking control noise to qubit state fidelity.

Their work, detailed in “Fidelity Relations in an Array of Neutral Atom Qubits – Experimental Validation of Control Noise”, utilises a ten by ten array of optically trapped rubidium-85 atoms, manipulated via microwave fields, to systematically analyse the impact of artificially introduced noise on quantum state preparation and maintenance. The results demonstrate strong agreement with predictions derived from the stochastic Schrödinger equation, offering valuable insight for noise characterisation and the development of robust control strategies in near-term quantum devices.

Research demonstrates a strong correlation between experimentally measured qubit fidelities and theoretical predictions derived from the Stochastic Schrödinger Equation, validating a model for understanding control noise in quantum systems. The study utilises a 10×10 optical tweezer array containing Rubidium-85 atoms as a qubit platform, manipulating hyperfine qubits with a global microwave field. Researchers actively introduce defined noise profiles into the control signals, enabling systematic investigation of noise impacts.

Three distinct noise models are analysed: white noise, representing random, uncorrelated fluctuations; Ornstein-Uhlenbeck noise, a correlated noise that reverts to a mean value and is more representative of physical systems than white noise; and Brownian motion, characterised by a 1/f² power spectral density, which describes a specific type of random movement. Mathematical modelling employs Stochastic Differential Equations (SDEs) to describe the evolution of the introduced noise, while the Stochastic Schrödinger Equation predicts qubit behaviour under these conditions. Qubit fidelity, a measure of the accuracy of quantum operations, serves as the primary metric for quantifying performance under varying noise conditions.

The research confirms the accuracy of the theoretical model in capturing the effects of control noise on qubit performance, providing insights into how different noise characteristics – randomness, correlation, and power spectrum – affect fidelity. This work establishes a robust framework for understanding and mitigating noise in quantum computations, particularly relevant for advancing Noisy Intermediate-Scale Quantum (NISQ) technologies.

Future research will extend the model to more complex quantum systems, explore novel noise mitigation techniques, and investigate the impact of different noise sources. Optimisation of quantum control pulses to minimise noise effects and the development of fault-tolerant quantum computers also form key areas of ongoing investigation. This research provides a valuable contribution to quantum information processing by offering a validated theoretical framework and experimental insights into the critical issue of qubit noise and its impact on fidelity.