QoE-Driven Optimization in 5G O-RAN-Enabled HetNets for Enhanced Video Service Quality
Bharat Agarwal, Mohammed Amine Togou, Marco Ruffini, Gabriel‐Miro Muntean
Abstract
Many innovative applications are projected to be supported by 5G networks across three verticals: enhanced mobile broadband, ultra-reliable low-latency communication, and massive machine-type communication. Given the constraints of the current radio access networks (RANs), accommodating all these applications, considering their quality of service and quality of experience (QoE) requirements is not practical. OpenRAN is a new architecture touted as the most viable next-generation RAN solution. It promotes a software-defined component, the RAN intelligent controller (RIC), which governs and supplies intelligence to optimize radio resource allocation, implement handovers, manage interference, and balance load between cells. RIC has two parts: non-real-time (non-RT) and near-RT. This article introduces a novel QoE enhancement function (QoE <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> F) xApp to enhance the functionality of a near-RT RIC through providing efficient resource provisioning to users requesting high-resolution video services. It deploys an innovative adaptive genetic algorithm to perform optimal user association along with resource and power allocation in HetNets. Simulation results demonstrate superior QoE <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> F xApp performance in terms of VMAF and MoS for two different resolution videos and diverse numbers of users.