The electromagnetic radiation prediction method of 5G base station is studied. A General Regression Neural Network(GRNN) model-based electromagnetic radiation environment representation method for base station is proposed, and the instantaneous broadband electric field strength at the ground plane of the theoretical maximum radiation point around the base station is predicted. 80% of the data is taken as the training set and 20% as the test set. Given the antenna transmission power, the distance between the 5G base station and its theoretical maximum radiant point, the data transmission time, the obtained Mean Absolute Percentage Error(MAPE) is 0.087 1, and the operating time is 3~5 min. The method shows good prediction accuracy and fast running speed. At the same time, by comparing with other models, the superiority of the method is verified, which is manifested in the substantial improvement of prediction accuracy and efficiency. With the increase of the prediction range around the base station, the advantages become more obvious, and the method has good applicability to different scenarios.