A Comprehensive Prediction Model for VHF Radio Wave Propagation by Integrating Entropy Weight Theory and Machine Learning Methods
Jian Wang, Yulong Hao, Cheng Yang
Abstract
In order to improve the prediction accuracy and robustness of radio wave propagation in the very high-frequency (VHF) band, we proposed a combination prediction model (CPM) based on the entropy weight theory and the machine learning method. The improved entropy weight method is used to determine the initial weights of recommendation ITU-R P.1546 and ITU-R P.2001 in the CPM based on the systematic analysis of the prediction performance of the two ITU-R models. Instead, we modeled the weight mapping using the machine learning method. The statistical comparison shows that the prediction accuracy and robustness of the proposed model are better than ITU-R P.1546 and P.2001 models, in which the root-mean-square error (RMSE) of path loss is reduced by 3.56 and 7.35 dB, respectively. In addition, the relative error (RE) of path loss is reduced by 0.0157 and 0.0255 dB, respectively, corresponding to a 40.53% and 52.62% reduction in the RE of path loss. The above research can provide basic support for the regional assimilation of radio propagation prediction methods.