ML-driven optimisation of physical layer characteristics in an interweaving of ICT and metaverse

Abstract:
By focusing on future disruptive technologies, systems related to metaverse and artificial intelligence (AI) seem to be at the forefront of the imagination of today’s high-end technical development. Although they promise optimisation in various quality attributes, there is a lack of a homogeneous scheme to describe the embedded system properties. To address this deficiency, we propose a taxonomy for describing ML-based optimisation of information and communications technology (ICT) that is interconnected with the metaverse. The scheme provides a solid basis for impact assessments, requirements definitions, as well as an essential foundation for trust and understanding the complex ML-based systems in the also very complex supersystem metaverse. Furthermore, the scheme creates a basis for the specification of requirements for the metaverse, which can be formulated for individual domains, e.g. health, energy or agriculture, in order to improve the explainability of the interconnectedness of technical and non-technical subsystems of the metaverse supersystem as well as the associated impacts.