Digital Twin Modeling for NV Magnetometry

Abstract

This thesis presents the development and application of a digital twin modeling framework for nitrogen-vacancy (NV) center-based magnetometry, advancing the field of quantum sensing. A surrogate model serves as a computational representation of the physical NV magnetometer system, enabling comprehensive exploration of parameter spaces to optimize device design. Leveraging machine learning techniques, this study optimizes control mechanisms, including the design of learned analog filters, to enhance system performance. This research investigates the fundamental limits of NV magnetometer performance, identifying strategies to minimize power requirements while maintaining high sensitivity. A dynamic framework is implemented to update the surrogate model’s parameters in real-time based on experimental measurements, ensuring accurate fidelity to the physical system. Additionally, the optimized control strategies are simulated within the digital twin environment, demonstrating their potential for advanced quantum sensing applications such as magnetocardiography (MCG) for heartbeat detection.