Assessment of Rain Attenuation in Satellite Telecommand Signals Using Numerical Weather Prediction Model and Artificial Neural Networks

In this study, the signal fading prediction model was implemented with artificial neural networks by obtaining precipitation parameters and atmospheric conditions in order to evaluate the signal attenuation observed in the telecommand signals of the geosynchronous telecommunication satellite. For a real situation, the application of artificial deep neural networks (ANN) over the 24-hour high-resolution Weather Research & Forecast (WRF) model simulation outputs was performed on two high-resolution simulation areas and used to estimate and model satellite signal attenuation. Preliminary results obtained from error analysis (RMSE) on multiple-input single-output feedforward neural network (MISO FFNN) prediction model outputs tested with various artificial neural network algorithms agree with observations of signal attenuation observed in downlink telemetry (receiver automatic gain control TM) obtained from the satellite receiver equipment. showed quite high correlation accuracy.