AI-assisted dual mode WDM system for anomaly detection and localization integrated communication and sensing

Ensuring the security and reliability of optical networks is critical for modern society. Distributed optical fiber sensing (DOFS) technology is an effective measure for managing and protecting optical infrastructure. To reduce the costs associated with sensing and maximize the utilization of channel resources, this paper proposes an integrated communication-sensing wavelength-division multiplexing (WDM) system featuring dual modes for anomaly detection and localization. By leveraging the assistance of the Long short-term memory (LSTM) algorithm and phase sensitive optical time-domain reflectometer (φ-OTDR), intelligent monitoring and precise localization of optical communication networks are achieved in two distinct modes. Furthermore, the dual-mode switching between detection and localization minimizes the need for additional communication time slots or bandwidth resources in DOFS system, thereby achieving efficient communication-sensing integration.