Machine learning-enhanced 3GPP channel modeling for 5G networks: A vendor-calibrated framework with cross-scenario validation

This study presents a regression-based framework integrating 3GPP TR 38.901 channel models with vendor-specific equipment parameters (Nokia, Huawei, ZTE) to predict 5G link performance across diverse scenarios (0.7–60 GHz). Findings indicate that ANN and decision tree models achieve high throughput accuracy, while mixed-scenario training is essential for model generalization across urban and rural environments.