An Optimal Framework for Integrating O-RAN and MEC via Pooled DU/CU Deployment

In this paper, we integrate the O-RAN environment with MEC resources through a comprehensive mixed-integer linear programming (MILP) formulation that jointly minimises deployment costs for DU/CU and application-layer workloads as well as transmission and processing delays. Our model incorporates the pooling of O-DU resources (pO-DU) under strict delay constraints, enabling efficient resource pooling at the edge to accommodate varying traffic demands and reduce operational and capital expenditures (OPEX and CAPEX). Moreover, the framework leverages the RAN Intelligent Controller (RIC) to orchestrate real-time control loops alongside MEC resources for computationally intensive tasks, ensuring that each network slice meets its specific capacity and latency requirements. Consequently, the system's intelligence is not limited to the RIC; we optimise offloading to MEC or cloud servers to satisfy latency and data processing requirements. Simulation results demonstrate that our solution effectively balances cost, latency, and resource utilisation, providing valuable insights for operators seeking to deploy next-generation networks with integrated O-RAN and MEC capabilities.