Research On Wireless Intelligent Propagation Model Based On Deep Learning
Min Wu, Changqing Li, Chenxiao Song · 2020
In order to better meet the growing and diversified market demand of users, mobile operators need to deploy a large number of base stations to improve the quality of radio wave propagation. Accurate and efficient network estimation is of great significance for accurate 5G network deployment. In this study, PCRR and RMSE of the reference signal receiving power (RSRP) of the observation points are taken as the optimization objectives, and the features are extracted from the perspective of model analysis and data analysis. Based on these features, a neural network is established for multi-objective learning and prediction of radio wave signal strength. Finally, the test data set of the radio wave signal strength prediction model based on deep learning achieved root mean square error of 8.59 and PCRR of 23.6%, which has a strong practical reference value for engineering practice.