Distributed Service Provisioning for Disaggregated 6G Network Infrastructures

6G Systems are expected to support a variety of services over a common infrastructure that is efficiently shared through slicing. Novel Quality of Experience (QoE) architectural models and resource assignment schemes are needed that are able a) to differentiate Service Data Flows (SDFs) originating from the same or multiple user equipment (UE), b) react to changes of the underlying physical infrastructure and c) scale with the number of connected devices. Currently, this functionality is provided by centralized management and network orchestration platforms suffering scalability issues. Thus, future systems are expected to operate in a distributed manner allowing applications to directly intervene in the relevant control processes to guarantee the required QoE. In response to this, the present study focuses on the development of a novel flow assignment scheme supporting applications running at the UEs. The scheme is based on Evolutionary Game Theory (EGT) and allows the application functions (AFs) responsible for the control of the UE applications to take traffic routing and steering decisions in a fully distributed way minimizing charging costs and maximizing their QoE. The charging function of the EGT model relies on profiling data extracted from an open-source 5G platform deployed in a practically implemented testbed environment.