AI-Driven Proactive Handover Optimization for NextG O-RAN Systems

Next-generation cellular networks are shifting towards cloud-based infrastructures emphasizing programmability, virtualization, and modular architectures. The open radio access network (O-RAN) paradigm is a promising approach to overcoming traditional RAN limitations by offering an open framework for data-driven optimization at the individual user level. However, the open and dynamic nature of the network introduces challenges such as fluctuating resource demand, varying signal quality, and uneven load distribution across base stations. To maintain consistent service quality and optimize network performance, intelligent handover mechanisms are essential, enabling dynamic reassignment of user connections to the most suitable base stations. To address these challenges, we propose an AI-enabled framework tailored for the O-RAN environment, aiming to optimize handover decisions while ensuring optimal user experience and efficient resource utilization. We formulate the handover optimization as a problem of maximizing system throughput within acceptable thresholds for other performance metrics. To solve this, we adopt a two-step approach. In the first step, an LSTM model deployed in the non-RT RIC forecasts key network metrics for the upcoming time slot. These predicted metrics are then forwarded to the near-RT RIC, where a genetic algorithm is employed to compute optimal handover decisions based on the forecasted conditions. Experimental results demonstrate that our framework accurately predicts network conditions and achieves up to $8 \%$ higher throughput compared to baseline methods, validating its effectiveness in dynamic and data-driven network environments.