Dynamic Phase Reference Synthesis Methodology for Resolving Magnetic Field Sign Ambiguity in Wireless Electromagnetic Tracking

Wireless electromagnetic tracking (WEMT) technique offers full six degrees-of-freedom (6-DoF) motion capture without the line-of-sight issue, making it highly suitable for wearable devices and virtual/augmented reality. However, WEMT faces a critical challenge: magnetic field sign ambiguity caused by the inability to calculate the synchronous phase difference between the transmitter and the wireless receiver, which leads to ambiguous solutions or restricts tracking to a single quadrant. Existing solutions typically rely on auxiliary hardware, such as inertial sensors or wireless communication modules, to resolve this ambiguity, increasing system complexity. To address this issue, this study proposes an auxiliary-free dynamic phase reference synthesis (DPRS) methodology based on a synthetic virtual transmitter (SVT), which dynamically generates a phase reference for each transmitting coil from receiver measurements using time-evolving sinusoidal models. Crucially, the SVT integrates real-time calibration based on dominant signal components to suppress phase drift, thereby ensuring long-term stability without wired synchronization. Numerical simulations validate the feasibility of the proposed methodology, demonstrating robustness even under varying signal-to-noise ratios. A long-term operation experiment lasting 1 hour was conducted on a self-developed prototype, demonstrating sustained performance with maximum pose errors of 0.48mm and 0.20°. Dynamic 4-quadrant experiments achieved root-mean-square errors of 6.61mm and 0.84°. The lightweight DPRS algorithm enables broader WEMT applications in human-machine interfaces and movement sciences by resolving magnetic field sign ambiguity.