Deliverable D3.3 Technologies, architecture and deployment advanced progress

This report provides the third update out of four planned for the 5G-TOURS project presenting overall work performed in WP3 (network architecture and deployment). The scope of this deliverable is to report the progress made in the 5G-TOURS network architecture for the 5G network deployment at an advanced stage within the three trial sites (Turin, Rennes, and Athens). This report focuses on the following aspects: • An update of the overall project network architecture based on 5G EVE, relevant standards components and 5G-TOURS innovations, as well as architecture instantiation per trial site; • The status of the network deployment on each trial site; • The technology evolutions conceived in the 5G- TOURS project to address the use cases and their implementation status. This document provides substantial updates in all the three areas, as discussed in the following. An update of the 5G Network Architecture is provided, resulting from the continuous effort devoted to the refinement of the network architecture. In particular, we consolidated the architectural view by: • describing the requirements coming from the NEST templates generated by WP2, • introducing the different technology evolutions and • defining the insertion strategies. Concerning network deployment: as the project transitions from phase 1 deployment to phase 2 deployment we report the new infrastructure that has been installed in the different sites. Most notably we remark, a better integration of the 5G-TOURS project into the 5G EVE platform. Technology evolution highlights 5G-TOURS work focused on a number of specific evolutions for the 5G network technologies: Enhanced MANO, Big Data and AI Orchestration, Broadcast and Service Layer. In this document, we present the instantiation of the novel technologies in the different sites, discussing how some of them are necessary enablers for many of the use cases that will be presented by the project. More specifically, 5G-TOURS provides solution for the Enhanced MANO by using also AI elements as well as a novel Broadcast Solution. Finally, all of them are made available to the verticals through the Service Layer. More specifically, 5G-TOURS developed solutions for the integration of MANO with specific AI algorithms, In this document, we mostly deal with two aspects that are related to: • the needed extensions of the baseline MANO architecture and implementation towards the integration of AI (e.g., the extension needed to support AI agents) and • the design and evaluation of specific AI algorithms into this architecture. The Service Layer, instead, is detailed into many flavours that specifically tackle an area of the vertical to network interactions. Among others, we specify implementation of the service layer for the direct interactions between the vertical and the AI governing the system (as also demonstrated in the ETSI ENI PoC developed by the 5G-TOURS partners and we cover the multicast/broadcast support in two axes: the development of the new network functions being specified at 3GPP Release 17 and the optimization of the transport delivery stack. Future work will be focusing on the finalization of the architecture, especially for the insertion over 5G EVE architecture, both functional and physical, with the deployment of the final infrastructure in each site. This will include also the integrations of the technology evolutions towards the final demonstrations of the UCs