A Multi-Dimensional Deep-Learning-Based Evaporation Duct Height Prediction Model Derived from MAGIC Data

The evaporation duct height (EDH) can reflect the main characteristics of near-surface meteorological environment, which is essential for designing a communication system under this propagation mechanism. This study proposes an EDH prediction network with multi-layer perception (MLP). Further, we construct multi-dimensional model (multilayer-MLP-EDH) first time by adding spatial and temporal “extra data” derived from measurements. experimental results show that: (1) compared naval-postgraduate-school (NPS) model, root-mean-square error (RMSE) meteorological-MLP-EDH reduced to 2.15 m, percentage improvement reached 54.00%; (2) parameters reduce RMSE 1.54 m 66.96%; (3) multilayer-MLP- match measurements well at both large small scales attaching extra height, further 1.05 77.51% NPS model. proposed significantly improve accuracy has great potential quality, reliability, efficiency ducting in ducts.