Fried Parameter-Based Cost Function for Dynamic Metropolitan Free-Space Optics Circuit Routing
Free-Space Optics (FSO) is a highly promising technology for high-capacity metropolitan (Metro) networks, yet its performance is severely degraded by atmospheric turbulence (AT). This phenomenon induces channel impairments that degrade the optical carrier and significantly increase the bit-error rate (BER), an effect further exacerbated in dense urban scenarios by local heat islands. To counteract these limitations, mesh Metro-FSO networks have the capability to circumvent critical AT-affected links by autonomously reconfiguring their routing nodes. To address this issue, this paper introduces an AT-aware dynamic circuit routing framework that leverages real-time environmental measurements from public meteorological stations. These data are used to dynamically estimate AT channel conditions, enabling greedy-based routing algorithms to compute optimal paths within a representative topology in Rio de Janeiro, Brazil. The framework is evaluated using real-world meteorological time series and a simulated urban heat-island (HI) stress scenario. The results demonstrate the feasibility of environmental data-driven routing and showcase a significant improvement in link reliability, reducing the cumulative BER by up to 93% under severe heat-island conditions compared to a static configuration. Ultimately, this work highlights how integrating real-time climate awareness into network-layer control can substantially enhance the resilience of urban optical wireless infrastructures.