Optimized Dynamical Decoupling in a Model Quantum Memory

We present experimental measurements on a model quantum system that demonstrate our ability to dramatically suppress qubit error rates by the application of optimized dynamical decoupling pulse sequences in a variety of experimentally relevant noise environments. We provide the first demonstration of an analytically derived pulse sequence developed by Uhrig, and find novel sequences through active, real-time experimental feedback. These new sequences are specially tailored to maximize error suppression without the need for a priori knowledge of the ambient noise environment. We compare these sequences against the Uhrig sequence, and the well established CPMG-style spin echo, demonstrating that our locally optimized pulse sequences outperform all others under test. Numerical simulations show that our locally optimized pulse sequences are capable of suppressing errors by orders of magnitude over other existing sequences. Our work includes the extension of a treatment to predict qubit decoherence under realistic conditions, including the use of finite-duration, square $\pi$ pulses, yielding strong agreement between experimental data and theory for arbitrary pulse sequences. These results demonstrate the robustness of qubit memory error suppression through dynamical decoupling techniques across a variety of qubit technologies.