Investigations into the suitability of artificial neural network for the pr
ediction of rain attenuation based on radio, meteorological and geographica
l data from ITU-R data bank are presented. First successful steps towards a
prediction model of rain attenuation for radio communication based on adap
tive learning from the measurement are made. Rain attenuation prediction wi
th the model based on artificial neural network shows good conformity with
the measurement. Moreover, a new evolutionary system, EPNet is used to evol
ve the artificial neural network rain attenuation model obtained both in ar
chitecture and weight, and an optimal rain attenuation model with simpler a
rchitecture and better prediction accuracy based on EPNet-evolved artificia
l neural network is obtained. Compared with the ITU-R model, the EPNet-evol
ved artificial neural network model of rain attenuation proposed in this pa
per improves the accuracy of rain attenuation prediction and creates a nove
l way to predict rain attenuation.