Simulation and Performance Evaluation of 5G Downlink Throughput over 3.5 GHz and 28 GHz Bands: An SNR-Based Analysis Using MATLAB

This paper presents a comprehensive simulation and performance evaluation of 5G downlink throughput over two prominent frequency bands, 3.5 GHz (sub-6 GHz) and 28 GHz (mmWave) using MATLAB R2023b and the 5G Toolbox. The simulation investigates the relationship between Signal-to-Noise Ratio (SNR) and downlink throughput, leveraging a detailed modelling of the 5G New Radio (NR) physical layer, including Orthogonal Frequency Division Multiplexing (OFDM), Low-Density Parity-Check (LDPC) coding, MIMO configurations (4×4 and 8×8), and advanced beamforming algorithms. The SNR is calculated from the received signal power and thermal noise, incorporating link budget parameters and 3GPP TR 38.901 channel models for Urban Macro (UMa) and Urban Micro (UMi) environments. Throughput is estimated using the Shannon capacity, and further refined using modulation and coding scheme (MCS) parameters, number of resource elements, and slot configurations. Path loss models specific to each environment and frequency band are applied, with significant differences observed due to increased propagation loss at 28 GHz. Simulation results reveal that the 28 GHz band achieves significantly higher peak throughput due to larger bandwidth and beamforming gain, despite suffering from higher path loss. The 3.5 GHz band, while offering lower peak rates, demonstrates more stable throughput across mobility scenarios due to better propagation characteristics. The throughput-SNR curves highlight the non-linear gain from higher-order MCS at high SNR regions, especially for 8×8 MIMO configurations. These findings underscore the importance of adaptive frequency and antenna strategies in 5G network planning and optimization.