Analyzing nuclear spins in diamond at room temperature based on traversal periodic-signal identification

Abstract

Detecting and imaging nuclear spins are of fundamental importance for spin-based quantum information processing in diamond. It is often realized by means of dynamical decoupling (DD) strategies, where a high-efficiency method for DD spectral analysis is required. Previously, a deep-learning-based algorithm is developed and applied on a cryogenic nitrogen-vacancy center experiment. Here, we improve the method by using a traversal periodic-signal identification approach before the deep-learning processing. With this improvement, low-resolution DD spectra with overlapped peaks could be well separated. This enables the deep-learning procedure being generally used in room-temperature experiments, where the measured spectra are often broadened with temperature. We apply this improved method in experiment, and its produced results match well with expectation. This method promises a wide range of applications in other spin-based systems.