XApp-Driven Multi-Metric Handover in 5G Open RAN with COTS User Equipment

In increasingly dense 5G network deployments, efficient handover (HO) mechanisms are essential to ensure service continuity, high user Quality of Experience, and optimal utilization of radio resources. Traditional HO strategies, typically based on static thresholds, are often unable to adapt to dynamic network conditions. In contrast, the Open RAN (O-RAN) architecture enables flexible and programmable control of Radio Access Network functionalities, supporting the development of adaptive, data-driven mobility management solutions. This work investigates the potential of xApp-based HO control within an O-RAN 5G New Radio (NR) environment. Leveraging the OpenAirInterface (OAI) platform and the FlexRIC framework, we design and implement a multi-metric xApp capable of monitoring key radio and network indicators in real time, including Key Performance Metrics (KPM). Based on these parameters, the xApp dynamically triggers N2 inter-frequency handovers through a weighted decision mechanism. The proposed solution is validated on a physical testbed with Commercial Off-The-Shelf (COTS) User Equipment, demonstrating reduced HO interruption times and improved responsiveness to varying network conditions. Furthermore, the study reveals device-specific behaviors in random access procedures that can significantly affect HO latency, highlighting the importance of real-world experimentation.