AI/ML-Based Sensing-Assisted Energy-Efficient Communications in Next-Gen Cellular Networks

5G networks promise to transform our technology experience by delivering ultra-high speeds and low latency, enabling applications like Augmented Reality (AR) and Connected Autonomous Vehicles (CAVs). However, 5G's higher frequencies reduce its range and lead to performance inconsistencies, especially for users on the move. Moreover, the energy consumption of 5G is significantly higher than its predecessor, 4G, raising sustainability concerns. In this paper, we explore a solution that combines the strengths of Integrated Sensing and Communication (ISAC) with the advanced analytics capabilities of the Network Data Analytics Function (NWDAF) in 5G networks. We leverage two new functions, Sensing Service Function (SSF) and Energy Efficiency Control Function (EECF), designed to work together to make smarter, more energy-efficient network decisions. By optimizing base station downlink transmit power, our approach not only reduces energy consumption but also carefully balances the trade-offs between latency and energy efficiency. Our findings suggest a promising path toward a greener and more reliable future for 5G and beyond networks.