Environment-Aware Real-Time Intelligent Central Unit-User Plane Migration in O-RAN

This paper introduces a dynamic migration framework for Central Unit User Planes (CU-UP) deployed on edge servers within Open Radio Access Network (O-RAN) environments. To bolster resilience at the edge, system and environmental metrics are continuously monitored. System metrics are collected via custom xApps running at the Near-Real-Time RAN Intelligent Controller (Near-RT RIC), while environmental data is gathered and stored in InfluxDB for real-time visualization and alerting through Grafana. When predefined thresholds are breached, the system initiates a seamless container migration of the CU-UP to a standby server using the Robot framework based Automation Platform (RAP). The proposed RAP approach achieves an average migration time of 1.45 seconds with an User Element Service Recovery Time (UE SRT) of approximately one second. The proposed framework significantly enhances Quality of Service (QoS), Quality of Experience (QoE), and overall network reliability in resource-constrained edge deployments. This combination of monitoring both system-level and environmental factors ensures that the CU-UP is migrated intelligently under stress from any circumstance.