Machine Learning Modeling for Radiofrequency Electromagnetic Fields (RF-EMF) Signals from mmWave 5G Signals

5G is the next-generation mobile communication technology expected to deliver more excellent data rates than Long Term Evolution (LTE). It offers ultra-low latency and ultra-high dependability, allowing revolutionary services across sectors. 5G mm-Wave base stations may emit harmful radiofrequency electromagnetic fields (RF-EMF). However, RF-EMF concern raises health and safety. From our research, the RF-EMF prediction model lacks papers or publications. This study uses IEEE and ICNIRP standards for assessment and exposure limits. The measuring campaign analyses one sector of a 5G base station (5G-BS) operation on 29.5 GHz in Cyberjaya, Malaysia. This study proposes two prediction models. The first model predicts signal beam RF-EMF, and the second predicts base station RF-EMF. Each model contains three neural network techniques to forecast RF-EMF values: Approximate-RBFNN, Exact-RBFNN, and GRNN. The results are analysed and compared with the measured data, observing which algorithm is more accurate by calculating the RMSE of each algorithm. As a result, it can observe that the Exact-RBFNN algorithm is the best algorithm to predict the RF-EMF because it shows a good agreement with the measured value. Besides that, in 1-min duration, the difference between predicted and measured values reached 0.2 less channels. However, in 6-min and 30-min, it can observe more accurate results since the differences between values reached 0.1in the situations. In addition, the ICNIRP standard was used and compared with the result and validation value of the algorithms.