On the design of a native Zero-touch 6G architecture

The complexity of envisioned 6G telecommunication networks requires an intrinsically intelligent architecture designed to autonomously adapt to dynamics with end-to-end zero-touch service automation operations. Motivated by this vision, this paper tries to formulate concepts and solution aspects towards designing a native Zero-touch 6G architecture. Our discussion concentrates around three main pillars, i.e. (i) introducing Machine Learning (ML) models in the core design of the 6G architecture as native functions rather than add-on model solutions; (ii) distributing 6G functionality to different components up to the extreme edge; to (iii) leverage technology leaps enabling, e.g., the use of multi-access technologies and peer-topeer communications besides the standard cellular connectivity and other centralised functionality