Rain Attenuation Prediction Models of 60GHz Based on Neural Network and Least Squares-Support Vector Machine

Abstract

Although 60 GHz mmWave (millimeter-wave) has attractive features and promising applications, it is affected seriously by rain attenuation. Based on the neural networks and SVM (support vector machine), two novel rain attenuation prediction models for 60 GHz millimeter-wave are proposed in this paper. We respectively applied the BP (back-propagation) neural network and LS-SVM (least squares-support vector machine) to simulate the non-linear relationship between rainfall intensity and rain attenuation, then the two models are compared with general ITU-R model. Experimental results showed that both of the proposed prediction models are indeed superior to the existing ITU-R model for rain attenuation prediction in the sense of both accuracy and stability while LS-SVM is the most promising model for the prediction of rain attenuation.