6G-INTENSE: Intent-driven native AI architecture supporting network-compute abstraction and sensing at the deep edge

The 6G ecosystem is anticipated to encompass a diverse array of business actors, including mobile network operators, infrastructure providers, and service providers. Existing network management systems, which traditionally map technological and administrative domains on a one-to-one basis, are inadequate for managing such a complex environment. This necessitates innovative distributed architectures with varying levels of abstraction. To address these challenges, we present 6GINTENSE,

a novel network management architecture designed to handle the distributed and heterogeneous nature of 6G through two key abstractions. First, the network-compute fabric component facilitates multi-domain mapping at the resource level by abstracting resource pools, including deep-edge nodes equipped with sensing capabilities. Second, the domain manager and orchestrator serves as a central entity for service orchestration,

providing a global perspective on the infrastructure. The proposed architecture leverages the intent-based networking paradigm at all architectural layers, emphasizing its critical role in enabling autonomous networks. Furthermore, it integrates an innovative native artificial intelligence toolkit, which is pervasive across all layers.