Geometry-Based Channel Modeling for Tunnel Entrance Scenarios of Mountain Railway

Railway communications are required to have the ability of constant connection to ensure railway operation efficiency and safety. However, most travelers have encountered instances of Internet disconnection or call dropping when the train enters and exits tunnels, significantly impacting the quality of train-ground communication. To achieve constant connection capability, it is crucial to characterize the channels in tunnel entrance scenarios for railway communication systems. This article provides a 3-D geometry-based stochastic model (GBSM) for tunnel entrance scenarios, which considers both the leaky coaxial cable (LCX) signals in tunnels and antenna signals out of tunnels. The channel statistical characteristics are derived based on the presented model, including time-frequency correlation function and delay power spectrum density. The validation using actual measurement data demonstrates that the proposed model may achieve an acceptable degree of accuracy. The boundary of the tunnel entrance scenarios is analyzed based on the numerical results. Furthermore, numerical results have shown that nonisotropic scattering degree and antenna position have substantial effects on channel correlation. These can be applied for forthcoming antenna deployment, network design, and performance evaluations of railway communications.