Optimized Dynamical Decoupling in a Model Quantum Memory

摘要

We present experimental measurements on a model quantum system thatdemonstrate our ability to dramatically suppress qubit error rates by theapplication of optimized dynamical decoupling pulse sequences in a variety ofexperimentally relevant noise environments. We provide the first demonstrationof an analytically derived pulse sequence developed by Uhrig, and find novelsequences through active, real-time experimental feedback. These new sequencesare specially tailored to maximize error suppression without the need for apriori knowledge of the ambient noise environment. We compare these sequencesagainst the Uhrig sequence, and the well established CPMG-style spin echo,demonstrating that our locally optimized pulse sequences outperform all othersunder test. Numerical simulations show that our locally optimized pulsesequences are capable of suppressing errors by orders of magnitude over otherexisting sequences. Our work includes the extension of a treatment to predictqubit decoherence under realistic conditions, including the use offinite-duration, square $\pi$ pulses, yielding strong agreement betweenexperimental data and theory for arbitrary pulse sequences. These resultsdemonstrate the robustness of qubit memory error suppression through dynamicaldecoupling techniques across a variety of qubit technologies.