Towards an intelligent 6G architecture: the case of jointly Optimised handover and Orchestration

Modern communication networks are intrinsically softwarised and programmable, hence essentially agile. However, critical functionality comes from legacy generations, limiting agility and posing performance and other considerations for demanding 5G services and - in the future - 6G services. This paper discusses issues with legacy functionality and envisions high-level intrinsically intelligent design toward a 6G architecture. We focus on the case of machine learning model-based network functionality that can jointly optimise both user handover and service resource orchestration, rather than engaging in costly and uncertain prediction-based actions to accommodate the stringent requirements of 6G services. Moreover, we discuss this paradigm as a design pilot for other intrinsically intelligent 6G network functions.