Long Term Prediction of Rain Rate and Attenuation using ANN and RNN Algorithms

Abstract

In the present study, rain rate and attenuation prediction model based on the Artificial Neural Networks (ANN) and the Recurrent Neural Network (RNN) - Long short-term memory (LSTM) is developed and the comparison results between the measured and prediction is shown for the developed models. The yearly and monthly variations of rain rate and rain attenuation at Ka and Q-bands are predicted for the period of July 2017-June 2018, from the measured dataset available for the period of July 2016 - June 2017 at a temperate location Vigo, Spain. The comparison results are shown in terms of 1) Complementary Cumulative Distribution Function 2) Root Mean Square Error (RMSE) and Standard Deviation (STD) of prediction errors.