DOI: https://doi.org/10.36347/sjet.2026.v14i05.008

Pages: 273-280

Multi-band LTE-5G communication networks are often utilized to serve various communication needs, such as high data rate, low latency, and connectivity. Nevertheless, evaluating the performance of such heterogeneous communication environments is challenging due to various factors, such as frequency band, traffic, and user mobility. This study presents an AI-driven framework for evaluating and optimizing radio network performance across low-band, mid-band, and mmWave frequencies. The proposed approach integrates system-level network modeling with data-driven prediction and adaptive resource allocation. A hybrid machine learning model, combining temporal and nonlinear learning techniques, is used to estimate key performance metrics such as throughput and latency. Simulation results show that the framework captures cross-band behavior and maintains consistent prediction accuracy under varying conditions. The adaptive mechanism improves resource distribution and reduces congestion compared to static methods. The findings demonstrate the effectiveness of integrating prediction and optimization within a unified system for multi-band LTE–5G environments.