Online reconstruction and tracking of optical spatial modes in ring-core fiber using self-guided state tomography

Mode characterization is indispensable for high-capacity spatial-division multiplexing systems, yet online monitoring of optical spatial modes in few-mode fibers remains a significant challenge. In this work, we propose and demonstrate online reconstruction and tracking of optical spatial modes in a ring-core fiber (RCF) using a self-guided state tomography (SGST) framework. By leveraging a reference-free stochastic optimization approach, we first benchmark the SGST method using high-dimensional free-space modes, followed by the reconstruction of complex mode fields in an RCF supporting seven modes across four mode groups. Evaluation of the reconstruction performance reveals that SGST achieves fidelities superior to off-axis holography and comparable to standard state tomography. Furthermore, we validate the robustness of the system under dynamic mode evolution induced by both active mode switching and passive fiber torsion. This stable and simple framework, with the capacity to track dynamically evolving states, offers a powerful tool for high-capacity optical communications and advanced fiber-optic sensing.