Analysis and Optimization of 5G Coverage Predictions Using a Beamforming Antenna Model and Real Drive Test Measurements

The ability to estimate radio coverage accurately is fundamental for planning and optimizing any wireless network, notably when a new generation, as the 5 th Generation (5G), in an early deployment phase. knowledge acquired from of previous generations must be revisited, particularly used path loss antennas models, 5G propagation intrinsically distinct. This paper analyses beamforming antenna model distinct models - 3 xmlns:xlink="http://www.w3.org/1999/xlink">rd Partnership Project (3GPP) Millimetre-Wave Based Mobile Radio Access Network Fifth Integrated Communications (mmMAGIC) applying them evaluate 3-Dimensional (3D) synthetic real scenarios, outdoor indoor environments. Further, Drive Tests (DTs) were 3GPP accuracy Urban Macro (UMa) scenarios. For model, it shown that use with multiple vertical beams advantageous Base Station (BS) placed below surrounding buildings; regular UMa surroundings, one beam provides adequate maximized after tilt optimization. exhibited Mean Absolute Error (MAE) 21.05 dB Line-of-Sight (LoS) 14.48 Non-Line-of-Sight (NLoS), compared measurements. After calibration, MAE LoS NLoS decreased 5.45 7.51 dB, respectively. Moreover, non-calibrated led overestimations user throughput up 25% 163%, respectively, calibrated predictions. Machine Learning (ML) algorithms resulted MAEs within range 4.58 5.38 LoS, 3.70 5.96 NLoS, Random Forest (RF) algorithm attaining lowest error.