Phase 3 Key Achievements 3.2

5G-VINNI:
5G-VINNI provided a 3GPP compliant mmWave 5G New Radio installation [5GVINNI-2] while it uses a testing automation supported by OpenTAP (released as open source). OpenTAP is a test automation solution supporting the fast and easy development and execution of automated tests [5GVINNI-3].

5G!Drones:
The 5G!Drones [5G!Drones-1] [5G!Drones-2] system architecture has been further refined [5G!Drones-3] and finally released with all its components including specific enablers developed [5G!Drones-4] [5G!Drones-5]to support the targeted trials. Its description indicates that the experimenter initially interacts via the web portal with the trial controller domain to describe the scenario, along with the technical description of the network slices and the target KPIs to be evaluated. The trial controller domain coordinates with the infrastructure controller domain so as to run the different experiments on top of the 5G facilities. Furthermore, in order to validate the trial and ensure safe access to the airspace by the drones, the trial controller domain also interacts with the U-space domain through a dedicated interface.

5G-CLARITY:
5G-CLARITY multiple wireless technology access scheme by proposing an enhanced access traffic steering, splitting and switching (enhanced-ATSSS) enabled by multi-point TCP (MPTCP) -based multiple access interfaces. [5G-CLARITY-1], [5G-CLARITY-2], [5G-CLARITY-3]

5G-COMPLETE:
5G-COMPLETE project promotes the use of mmWave mesh nodes enabling P2MP topologies and sub-THz radio systems allowing for fiber-like wireless connectivity [5GC-1]. The mmWave mesh nodes operate in the 60 GHz spectrum band and they use antennas with 360° total coverage, connecting in this way multiple low-cost client nodes which are used as the endpoints of the wireless transport network [5GC-2]. High-speed wireless transceivers targeting 100 Gbps soon have bee also proposed within 5G-COMPLETE [5GC-3]. The targeted carrier frequency is 240 GHz, and the parallel sequence spread spectrum (PSSS) is proposed as a modulation scheme. These high capacity, SDN-enabled THz nodes will be demonstrated in the upcoming experiments in Athens Demo Zone [5GC-2].

5GZORRO:
5GZORRO developed mechanisms for dynamic spectrum trading, allocation and sharing enabled by the multi-stakeholder marketplace. Specifically, 5GZORRO uses digital spectrum certificates which allow the creation of spectrum resource offerings in the marketplace and its utilization through the 5GZORRO framework by means of spectokens. Indeed, spectrum resources are digitalised in the marketplace DLT through the introduction of spectokens. The spectoken helps keeping track of the historical of transactions of the spectrum resource; and the use of the spectrum resource since the very first moment a radio resource made use of it. In the 5GZORRO marketplace, Spectrum Resource Providers acquire primitive spectokens from a Regulator stakeholder according to the accepted spectrum certificates and are allowed to create spectrum offers associated to spectokens derived from them (Derivative Spectoken). [5GZORRO-1][5GZORRO-2][5GZORRO-3].

ARIADNE:
The ARIADNE project designed and developed a complete radio access system, including antennas and RIS prototypes, to be integrated and demonstrated towards the project end.

AFFORDABLE5G:
In Affordable5G [AFFORDABLE5G-1], advanced deep reinforcement learning algorithms were designed and deployed based on O-RAN specifications for private network optimization. A cross-layer network telemetry architecture was also deployed in the same context to ensure end-to-end data collection and real-time analytics [AFFORDABLE5G-2].

DRAGON:
D-band transceiver front-end building blocks through a low-cost SiGe BiCMOS process, an active antenna array and advanced integration technology have been developed in the DRAGON.

6G BRAINS:
Sub-6 GHz, mmWave, sub-THz, and Optical Wireless Communications Fused Ray-Tracing Model and its Verification from Measurements [6GBRAINS-4]. A precise ray-tracing (RT) model from a real industrial scenario was generated from point cloud data obtained from extensive 3D laser scans. This model was calibrated, verified, and validated with simultaneous multi-band RF measurements at sub-6 GHz and two mmWave bands [6GBRAINS-1]. The ultimate goal is to obtain a fused model covering from sub-6 GHz to the optical spectrum, enabling the RF simulation of heterogeneous networks. This digital twin of the environment will be used to perform realistic analysis on different algorithms for sensing, localization, multi-band and sensor aided beam-forming, between others.

DAEMON:
DAEMON is contributing substantially to Open and Virtualized RANs, with a number of contributions to the O-RAN Alliance in WG 2 and WG 3 [DAEMON-2], research work on ML models for RAN Intelligent Control [DAEMON-3], and novel virtualization techniques [DAEMON-4].

Hexa-X:
Hexa-X project are working on technical enablers for Sub-THz communication, e.g., HW-impairment modelling, waveforms, beamforming and channel modelling as well as techniques for distributed MIMO for mmWave communication. Furthermore, Hexa-X is developing techniques for AI-based air interface design, counteracting HW impairments or improving spectral efficiency. Hexa-X is also exploring solutions for flexible networks looking at architectural enablers to improve and incorporate novel accesses such as D2D, mesh, NTN, and mesh networks.

MARSAL:
MARSAL completed the specifications of a cell-free based RAN, while also dynamic RU cluster formation algorithms and precoding optimization techniques for cell-free networks have been developed. The proposed RAN solution follows the specifications of O-RAN, which are extended from MARSAL’s work in order to include the cell-free approach.

REINDEER:
REINDEER develops ‘RadioWeaves’ technology, providing a new smart connect-compute platform consisting of interconnected distributed resources to provide uniform good service levels. Architectures are designed to support communication, localisation, and wireless power transfer to interact with energy-neutral devices. A new terminology and federation-based approach is proposed to allocate infrastructure resources to services. Channel characterisation for these new radio networks has been performed based on measurement campaigns.  [REINDEER-1] – [REINDEER-2] – [REINDEER-3].

5GRAIL:
5GRAIL will test the first railway application over FRMCS prototypes, via 5G n8 and n78 radio modules. We have obtained a FRMCS 1900 MHz prototype (1900 – 1910 MHz, TDD, 31 dBm), compatible ECC (20) 02.

5G-COMPLETE:
Low-cost high-capacity optical fronthaul solutions enabled by advanced modulation formats and wavelength-agnostic passive wavelength division multiplexing (WDM) technology have been delivered wihin 5G-COMPLETE. To meet the time-sensitive networking demands in support of 5G fronthaul, an FPGA-based implementation providing low latency and low packet delay variation following the latest IEEE 802.1CM specification has been presented [5GC-4]. Aside from the high-speed transport connectivity, the introduction of an intelligent access network equipment capable of hosting Mobile Edge Computing capabilities in a convergence scenario of PtP and PtMP topologies has been proposed [5GC-5]. Beyond the standard digital X-haul interfaces, 5G-COMPLETE has demonstrated a multi-technology hybrid transport architecture that comprises both analog and digital-Radio over Fiber (RoF) mobile network segments relying on a dynamically reconfigurable optical switching node [5GC-6].

TERAWAY:
TERAWAY has studied 5G evolutionary and beyond 5G access scenarios including densification, mmWave RAN and disaggregation to analyze and define the main application scenarios and use cases for THz-band wireless transport [TERAWAY-1]. B5G xHaul constitutes the main scenario for TERAWAY [TERAWAY-2] and requirements for THz transport in fronthaul / midhaul and backhaul architectures have been defined for fixed and moving node cases. System specifications have been defined to meet throughput requirements expected in evolutionary centralized and distributed radio architectures and to allow integration in SDN management architecture. Multi-band photonic-based W/D/THz tuneable communication modules have been designed and will be used in the project demonstrations, focused on fixed xHaul applications. Market analysis including 5G deployment timeline, mmWave product roadmaps and spectrum and network/service evolution has been conducted defining a preliminary roadmap for the use of mmWave/THz technologies in 5G/B5G mobile networks and the base for exploitation analysis.

Int5Gent:
Int5Gent [I5G-1] integrates a number of data plane technologies under a common service and network resource orchestration platform [I5G-2]. For the access segment the novel Sigma-Delta Radio over Fibre (RoF) scheme is defined and implemented [I5G-3] alongside the common digital RoF scheme and the forward looking Analogue RoF technology. A passive wavelength routing scheme is deployed for the fronthaul/metrohaul segment [I5G-4] and examined in parallel to mmWave mesh radio connectivity technology. A baseband processor platform for Int5Gent Edge Nodes is designed and implemented and supported by NIC-enabled synchronization engine for time-sensitive edge networking. Interfacing schemes are implemented for the integration of the various technologies as well as for the integration of the upper management and control plane.

B5G-OPEN:
B5G-OPEN is developing next generation metro-haul and core infrastructures based on innovative multi-band technology. [B5G-OPEN-1/2/3]

DAEMON:
One of the key research tasks of DAEMON consists of the design and development of ML-based network intelligence for in-backhaul intelligence. This consists of augmenting backhaul nodes with smart functionality to process traffic as it flows from the network to users or vice-versa by using with light and fast ML [DAEMON-4].

5G EVE:
Interworking framework to unify the end-to-end management of experiments over a multi-site 5G network facility composed of heterogeneous 5G site facilities (in terms of technologies, architectures, orchestration, metrics and tools) managed by different administrative entities ([5GEVE-5]). Therefore, the framework grants vertical users of 5G EVE platform service with a homogenous way of setting their experiments across the constellation of site facilities of 5G EVE ecosystem, and supporting both single-site and multi-site use cases ([5GEVE-3]).

5G-SOLUTIONS:
Cross Domain Service Orchestrator (CDSO) enabled end-to-end integration of use cases from 4 domains with 3 geographically distributed ICT-17 facilities and some stand-alone nodes and in some cases with a ZTA feedback loop. Challenges include Network access, Authentication, API adherence, payload & timing. Zero-touch automation (ZTA) mechanisms and solution for network and application optimisations, which allows the definition of rules specific to the 5G vertical, can be used in conjunction with Machine Learning methods to provide for proactive and adaptable behaviour. Full chain integration (including 5G platform VNF orchestration solution, KPI monitoring and collection systems from the platforms, applications, and KPI Visualization System, ML modules, as well as the CDSO) and closed loop optimisations have been proven in the context of the 5G-SOLUTIONS media use cases. Details in [5G-SOLUTIONS-3] … [5G-SOLUTIONS-5] The 5G-COMPLETE solution employs service-driven slice management in 5G and beyond 5G (B5G) networks, while integrating the orchestration of radio, transport, and core network resources as well as cloud and edge computing resources [5GC-7].

5G-COMPLETE:
5G-COMPLETE aims to build upon a well-defined architecture as well as the network management and orchestration layer with two main goals: i) develop novel networking and compute blocks, integrating them in a heterogeneous infrastructure where service-driven slice management can assist use cases in realistic deployment options [5GC-2], ii) develop solutions to provide a higher and more intelligent level of network management boosted by Artificial Intelligence/Machine Learning (AI/ML) assisted tools for improving operational efficiency, energy efficiency and reducing the operational expenditure of 5G and B5G networks [5GC-8].

5GZORRO:
5GZORRO defined and implemented mechanisms for intelligent and automated cross-domain slices and services management. Specifically, Intelligent and Automated Slice and Service Management (ISSM) functionalities focus on the automation of secure cross-domain business transactions enforcement, interacting with MANO tools slices and services deployment while optimizing resource allocation to inter-domain slices according to business SLAs. In particular, a workflow manager executes orchestration workflows in the context of a business transaction, such as extending a slices across domains in cooperation with Network Slice and Service Orchestration and MANO tools. On top of that, specific AI-driven optimization functionalities optimize the cost-efficiency trade-off of network services and slices required to be created in a context of a specific business transaction and continuously optimize services and slices that have been already set up in previous transaction flow executions. [5GZORRO-4]

ARIADNE:
In order to manage and orchestrate the complex D band systems, including the RIS, the ARIADNE project is applying AI/ML techniques and is preparing a corresponding demonstrator.

INSPIRE-5Gplus:
INSPIRE-5Gplus [INSPIRE-5Gplus-1] defined a High Level Architecture (HLA) of a zero-touch end-to-end smart network and service security management framework that enables not only protection but also addresses trustworthiness and liability in managing 5G network infrastructures across multiple domains [INSPIRE-5Gplus-2]. Based on the HLA, INSPIRE-5Gplus developed a set of enablers for the automatic and autonomic end-to-end and multi-domain security management by security policies and SSLAs and to optimize the orchestration, the provisioning, and the chaining of virtualised security functions, micro-services, and virtualised network functions. A detailed description of these enablers is described in D3.2 [INSPIRE-5Gplus-3].

LOCUS:
LOCUS put forth the integration and management of virtualized localization data collectors and processors into a virtualization platform to implement the LOCUS localization and analytics “as a service” model. LOCUS defined a high-level functional architecture, together with functions and components. The System Architecture and related Services were designed along with exploring initial deployment options and mapping to Standards. LOCUS contributed to the 3GPP working groups RAN and SA from concept development and solutions for Rel-17 to work item on NR positioning enhancement in Rel-18 and initial discussions in Rel-19. Use cases have been defined and proof of concepts with experimentations of localization techniques and location-based analytics have been conducted.

MonB5G:
Description of AI-driven and decentralised management and orchestration architecture, including security, of network slices, achieving the concept of Zero-touch Service Management (ZSM) [Monb5g-1], [Monb5g-2]. Description of AI-driven In-Slice Management (ISM) concept, which reduces number of external slice interfaces and separates slices’ management plane. Description of multi-domain orchestration framework, which provides a strong separation between orchestration domains.

TERAWAY:
TERAWAY provided the first prototype of a network slice manager that builds end-to-end network slices. In TERAWAY the Transport slice manager (TS-NSMS) includes SDN technology for interacting with fixed network switches and mmWave or THzWave radio modules to allocate and manage the resources required for building the network slices. TERAWAY has followed 3GPP, ONF and IETF specifications to deliver a transport controller to be incorporated into the 5G architecture.

5G-IANA:
The 5G-IANA Automotive Open Experimental Platform (AOEP) integrates different MANO frameworks for enabling the deployment of the end-to-end network services across different network segments as a virtualized infrastructure (vehicles, road infrastructure, MEC nodes and cloud resources). In fact, 5G-IANA will enable “on-vehicle” MANO by developing a programmable OBU/RSU ready to host containerized services. This will allow developers to take full advantage of the capabilities offered by the different nodes in the infrastructure (i.e., OBUs attached on vehicles and RSUs deployed on the roads) and to address an efficient and effective NetApp deployment and orchestration [5GIANA-1] – [5GIANA-2]. Additionally, 5G-IANA provides support for the orchestration of (distributed) Machine Learning (ML) services, with a particular focus on federated learning. The visibility of mobile, “on-vehicle” resources enable the support of DML orchestration primitives such as FL client selection [5GIANA-3].

5GMediaHUB:
5GMediaHUB completed the design of a Cross-Domain Service Orchestrator (CDSO) which is responsible for the orchestration of media NetApps. The workflows for NetApp onboarding, E2E slice ordering as well as interactions with testbeds’ 5G SOs and Domain Monitoring Systems (DMSs) are captured. [5GMediaHUB-2]

Smart5Grid:
Smart5Grid will integrate an M&O framework that will be responsible for managing and coordinating the operations and lifecycle of the virtualised communication/storage/computing resources of the telecommunications network, VNFs, services and NetApps. In addition, the M&O framework will be responsible for the coordination of the network slice over which the communication services and NetApps are instantiated.

VITAL-5G:
VITAL-5G has developed a platform ([VITAL5G-2], [VITAL5G-3]) to automate the provisioning of 5G-enabled T&L services over different 5G testbeds ([VITAL5G-4]) featuring NFV orchestration, network slicing and monitoring capabilities. NetApps, developed as cloud-native applications, are orchestrated in VM-based or container-based virtual infrastructures. The related network slices are instantiated or configured dynamically to meet the NetApps’ mobile connectivity requirements. Network and service KPIs are collected for experimental verification and performance validation, enabling diagnostics and closed-loop mechanisms for automated network and service re-configurations.

6G BRAINS:
Intent-based networking for B5G/6G industrial scenarios [6GBRAINS-5]. The concept of Intent-Based Network (IBN) emerged to introduce a layer of Artificial Intelligence (AI) in the sixth generation (6G) networks, solving the problems of traditional networks in terms of efficiency, flexibility and security. This technology has revolutionised the way we interact with systems by starting to communicate through intents. An intent is an expression of the desired state that you want to be realised and can be considered: Portable – can be moved between the different controller and network implementations and remain valid; Abstract – must not contain any details of a specific network;  The advantage of an Intent is flexibility, as it allows users to express policies using concepts and terminology that are familiar to the user without having specific knowledge in the field.

AI@EDGE:
The AI@EDGE project introduced the concept of Artificial Intelligence Function or AIF. An AIF is a component of an AI-enabled application deployed across the network from the extreme edge to a centralized core. The AIF models encapsulate all the aspects of the computation associated with the AI application including its requirements. This includes conventional requirements like CPU, memory, and storage as well as more AI-focused requirements like hardware acceleration, support for federated learning, etc. The AIF model explicitly accounts for the training and inference phases allowing the orchestration layer to understand if the reliability of an AI model is deteriorating and if a re-training should be triggered. This model has been tested to deploy and manage several federated learning-based AI applications [AI@EDGE-2]. The AIF concept is combined with a reusable ML pipeline model which allows ML models to be centrally managed and monitored by the AI@EDGE platform. [AI@EDGE-1] – [AI@EDGE-3].

B5G-OPEN:
B5G-OPEN is developing next-generation AI-empowered SDN Control and Orchestration system for next-generation multi-band optical networks. [B5G-OPEN-1/2/4/5]

DAEMON:
DAEMON is designing a Network Intelligence (NI) plane able to orchestrate intelligence across different domains (radio, backhaul, core) and layers (network function orchestration, network control, data-plane) [DAEMON-5].

DEDICAT 6G:
A second version of DEDICAT 6G architecture has been defined. The architecture work follows a precise and logical methodology that encompasses various steps, starting from an extensive requirement engineering process including a few new requirements from D2.3 [DEDICAT6G-2], followed by a revised functional decomposition that gives a full catalog at the functionalities which must be specified and implemented in order to fulfill those project technical objectives and an updated description of the non-functional properties it must implement. A logical pathway leads project from the requirement collection and analysis to the targeted system functional decomposition, which consists of a large set of components interacting with each other in order to implement the main concepts introduced in DEDICAT 6G project, namely 1) dynamic radio coverage extension using a variety of mobile access points and edge nodes, 2) dynamic intelligence distribution relying on those edge nodes and 3) federated learning-based trust management.  [DEDICAT6G-3]

MARSAL:
MARSAL has defined its virtual elastic infrastructure concept for the network management procedures, targeting the disaggregation of containerized application functions both horizontally (i.e., across edge sites) and vertically (i.e., from the cell site towards the core cloud).

TeraFlow:
Integration with telco cloud and MEC has been tackled in [TF-2] by considering the necessary interactions between an externally-selected NFV Orchestrator and the TeraFlowSDN at the time of automatically deploying network services embracing both compute (e.g., VNFs or CNFs) and networking (i.e., wide-area network) resources. Specifically, the NFV Orchestrator and the TeraFlow OS interact according to a client-server relationship. The selected NFV Orchestrator is the ETSI OpenSource MANO (OSM).

5GMED:
In cross-border scenarios, a UE using a virtualized service might need to change its PLMN when crossing the border, which requires provision of resources and configuration of the same service in the visited PLMN. 5GMED will design and implement a cross-operator service orchestration platform that enables MNOs, neutral hosts, and road/railways operators to deliver service continuity to end-users. The orchestration platform of 5GMED will federate the instances of two domain orchestrators following the specifications of the Operator Platform Group (OPG) [5GMED-1] – [5GMED-2]

Hexa-X:
The Hexa-X project is developing techniques to enhance the network performance and automation, by enabling incorporation of AI/ML. A key feature to enable this is network and device programmability which will enable frequent and customized configurations with a CI/CD approach. Furthermore, with a cloud-native architecture, dynamic function placements can allow tailored network operations in a small area or during a short period of time.

TeraFlow:
TeraFlowSDN has been implemented following the cloud-native architecture concept. The software is divided into micro-services that execute specific tasks and collaborate through messages to achieve the end goal for which the software is designed. In TeraFlowSDN, the communication between components is based on gRPC. We defined different RPC methods and messages and grouped them by micro-service [TF-1].

LOCUS:
LOCUS has developed an end-to-end service-based architecture with built-in security and privacy, that takes into account the challenges of handling sensitive data.

MonB5G:
Description of AI-based closed control loop framework to detect and mitigate attacks on network slices. Introduction of the Security orchestration and Security as a Service concept. Trusted architecture for network slicing deployment. Demonstration of the capabilities of the closed control loop to detect and mitigate attacks using three use cases: In-slice mMTC DdoS attacks on AMF [Monb5g-3] [Monb5g-4], aLTEr attacks (traffic steering and VNF instantiation), and Attack detection and mitigation on Federated Learning (FL) training process.

5GMETA:
5GMETA will permit to have a secure and private pipeline to manage services using 5G Network.

COREnect:
COREnect identified several checks and balances to realistically reduce Europe’s dependence on American and Asian technologies. COREnect also made recommendations to ensure trustworthiness and security even when certain components are of untrusted source, e.g., micro-kernel-based operating systems or meta-level fabric for multiprocessor system-on-a-chip design [COREnect-1, COREnect-3].

MARSAL:
MARSAL has developed a decentralized framework for confidentiality and hardware-accelerated security mechanisms, targeting safe exchange of confidential data among parties in a B5G/6G network, as well as the development of smart contracts among tenants requesting networks slides. In parallel, MARSAL has also developed a policy-driven solution for data security and privacy in multi-tenant environments.

TeraFlow:
TeraFlow has contributed with a centralised Cybersecurity Component have been implemented. The preliminary performance and scalability results of the modules using a supervised and an unsupervised learning model have been detailed in [TF-3]. The fundamental procedures of the Distributed Cybersecurity Component have also been implemented and the modules composing the component have been validated. Furthermore, the Distributed Ledger Technology (DLT) Component has been implemented based on the modular architecture of Hyperledger Fabric.

INSPIRE-5Gplus:
INSPIRE-5Gplus has explored and developed a set of security enablement having the potential to significantly contribute to 5G security evolution, such as Trusted Execution Environments, Distributed Ledger Technologies, Liability and Root Cause Analysis, enablers related to network automation & zero-touch management, SSLAs, and Multi-Domain security policies management. The consortium has also worked on significant extensions to the prediction, protection, detection, and mitigation closed loop with new paradigms using AI, ML, and data analytics such as DDoS detection using AI, proactive defence using MTD (Moving Target Defence), ML-driven binary vulnerability detection (for VNFs) as well as detection of malicious VM or container, investigation of possible adversarial Attacks against AI/ML techniques, and the design of the decision engine based on AI that relies on data analytics and an intelligent automated security policy orchestration. All these enablers are detailed in WP3 and WP4 deliverables [INSPIRE-5Gplus-3].

5GENESIS:
5GENESIS has provided a definition and implementation of the functional architecture for experimentation on 5G infrastructures. That architecture encompasses automation in the experimentation process, openness on VNF on-boarding procedures, and flexibility in KPIs monitoring [5GENESIS-2], [5GENESIS-3]. The experiment automation part of the 5GENESIS architecture has been released under the term Open5GENESIS Suite, which provide in an open-source manner the all the necessary components and tools to any communication platform to become subject of automatic tests and measurement campaigns [5GENESIS-5].

5GASP:
5GASP methodology and tools have already been developed and deployed in 6 testbeds across Europe, open for automated testing using the 5GASP methodology. A service portal provides a 1 stop interface to deploy NetApps and Tests with an ultimate goal of achieving a 5GASP Certification or their NetApps. [5GASP-1] – [5GASP-4], 5GASP provides all its tools using OSS licenses fostering the development of new NetApps [5GASP-5]. Furthermore, has provided educational material such as videos, manuals and an open forum [5GASP-6]

5G-ERA:
5G-ERA middleware and communication interface for robots. Robot applications are optimised for local deployment with limited network scalability. As a de facto standard in the robot community, ROS1 communication is restricted to broadcasting within the same network. ROS2, which is the latest improvement of the ecosystem addresses the limitation through multicast. Unfortunately, neither of them can satisfy the end-to-end multi-domains and multi-administration nature of real-world networking requirements. 5G-ERA middleware and “unified communication interface” are designed for large scale robot deployment under 5G environments. The interface maintains the support of legacy ROS communication paradigm, at the same time optimised for service endpoints via restful communication. The innovation on platforms and NetApps ensure a smooth switch over from local Wi-Fi based robot applications to global distributed 5G connected intelligence. [5G-ERA-1] – [5G-ERA-5].

5G-IANA:
5G-IANA will specify and implement an Automotive Open Experimental platform (AOEP) that provides the compute and communication/transport infrastructure as well as the management and orchestration components, coupled with an enhanced NetApp (Network Application) Toolkit tailored to the Automotive sector. The 5G-IANA NetApp toolkit will be linked with a new Automotive VNFs Repository including an extended list of ready-to-use open accessible Automotive-related VNFs and NetApp templates, forming a repository for SMEs to use and develop new applications.

5GInduce:
5GInduce will develop and integrate a NetApp management platform; intelligent OSS will be deployed; cloud orchestration platform will be based on scalable microservices; advanced user interface for porting of NetApps will be implemented.

5GMediaHUB:
Definitions of internal design and Northbound APIs for NetApps targeting the media applications (mCDN, 360 VR, Remote Broadcasting, UGC, Second Screens). 5GMediaHUB adopts the ETSI NFV model for NetApps and addresses cross-domain orchestration and slicing aspects. Moreover, a NetApps repository is implemented, offering asset management and catalogue services [5GMediaHUB-3].

EVOLVED-5G:
EVOLVED-5G has released a Workspace on NetApp development. It includes Cloud-driven SDK/CLI [EVOLVED5G-2] tools that come with a detailed “how to develop a NetApp” [EVOLVED5G-3] instructions to facilitate third party developers (including the development teams of each one of the EVOLVED-5G SMEs) to efficiently develop Network Apps. EVOLVED-5G implemented the 3GPP Common API Framework (CAPIF) following the specifications of 3GPP TS 23.222 and 29.222 Rel 17. The related CAPIF services [EVOLVED5G-4] are available in the project’s GitHub repository. EVOLVED-5G implemented the Network Exposure Function (NEF) Northbound standardised APIs following the specifications of 3GPP TS 29.522 Rel. 17 that realizes the openness of the 5G network core. The related standalone NEF emulator [EVOLVED5G-5] is available in the project github [EVOLVED5G-2]. EVOLVED-5G provides also an integrated NEF/CAPIF implementation, offering the whole experience of the 5G core openess to any interested third party.

Smart5Grid:
Smart5Grid will develop and integrate a platform to provide a common place for application developers and consumers. The Smart5Grid platform will be divided into three layers. The first layer will contain a NetApp Verification and Validation framework, an Open service Repository for these NetApps and an Interface Platform. The second layer will contain the virtualisation and telecommunications infrastructure with its associated management functions and orchestration functions. The third and final layer will contain the power infrastructure that contains the network components that connect to the NetApp services.

VITAL-5G:
VITAL-5G has defined a template for NetApp blueprints ([VITAL5G-3]) and implemented a platform ([VITAL5G-5]) for onboarding, provisioning, monitoring and experimentally validating NetApp-based T&L services. VITAL-5G platform provides a service catalogue for onboarding, sharing, and composing NetApp packages, and a portal for building, instantiating, and testing new NetApps and services, with mechanisms for automatic configuration of network slices, collection and analysis of network and service KPIs. VITAL-5G platform is open for third-parties. SMEs and research centres, acting as NetApp developers, can easily build new 5G-enabled applications for the T&L sector starting from the examples available in the catalogue and integrating application components from various vendors. The opportunity to experimentally validate NetApps in realistic and customizable 5G environments speeds up the whole development process. In this direction, VITAL-5G has defined a trials methodology ([VITAL5G-6]), with tools, procedures and testcase templates for experiments in T&L facilities extended with 5G infrastructures.

Int5Gent:
In order to fulfil its main objective, Int5Gent have defined and implemented a common end-to-end orchestration platform that extends from the end-user-defined service layer to the management of the transport and data plane resources [I5G-5]. The platform combines a) a vertical application orchestrator responsible for the service creation, onboarding and runtime management of services, b) a Management and Orchestration (MANO) framework, including features for 5G network slicing and service orchestration over a hybrid cloud/edge infrastructure and c) an SDN-based control plane, fully integrated in the Int5Gent MANO framework, for service-driven network configuration, runtime self-healing and self-optimization. Enhanced interfaces between NFV MANO elements, SDN-based WIMs for integrated Fiber-Wireless fronthaul/midhaul and cloud/edge infrastructures to enable network slicing in multi-domain and multi-technology environments are also designed and implemented.

5G-ROUTES:
5G-ROUTES implements CAM Services platform for offering technological enablers for cross-border CAM Use Cases.

LOCUS:
LOCUS has developed innovative signal processing techniques to enhance the 5G and beyond localization accuracy by also exploiting heterogeneous technologies such as multi-RAT (integration with GNSS, WIFi, Bluetooth, and UWB), multi-carrier, multi-connectivity, mmWaves, intelligent surfaces, and device-free localization. LOCUS has designed analytical methods (prediction and multimodal data processing) for the use of localization information in network management, as well as developed advanced analytic mechanisms to analyse the behaviour of devices and targets.

Hexa-X:
The Hexa-X project are exploring techniques for joint communication and sensing, to enable sensing of objects and environments employing the same hardware and infrastructure used for communication. Hexa-X is also exploring methods to incorporate the sensing and localization information into the network management to enhance the network performance, e.g., through location-based beamforming or trajectory-based connectivity optimization.

REINDEER:
REINDEER has developed algorithms and architectures for supporting interactive applications perceived as ‘real-time’ and ‘real-space’, i.e., where the physical and virtual worlds share the same reference frame. These tackle the performance/complexity/interconnect challenges in distributed deployment and cell-free operation. Protocols for initial access, synchronization, and calibration in new distributed networks, are designed. [REINDEER-2] – [REINDEER-4].

5G-SOLUTIONS:
Within the scope of 5G-SOLUTIONS project, an innovative smart KPI visualization system has been designed and implemented, leveraging cognitive machine learning techniques, big data analytics and cloud computing approaches, for facilitating and automating the presentation, benchmarking and performance validation of 5G network and application level KPIs against specific target 5G values. The KPIs are retrieved from the applications of 20 use-cases and their underlying 5G networks, which include ICT-17 facilities 5G-EVE, 5G-VINNI Norway, 5G-VINNI Patras, along with some standalone 5G nodes. Data harmonization was done to support the ZTA mechanisms within the project. [5G-SOLUTIONS-3], [5G-SOLUTIONS-6]

MoNB5G:
Definition of novel E2E slice KPIs for monitoring performance of slices. Graph-based learning for slice KPI prediction. Federated Learning for low SLA violations in beyond 5G network slicing. AI-based Intra and inter slice admission control. [Monb5g-5]. MoNB5G energy techniques: decentralised cross-domains Energy Efficient Decision Engine. Energy efficient at RAN and Edge. [Monb5g-6]. Publication of 5G Datasets collected from project testbeds contributing to the future research of the community [Monb5g-7].

TERAWAY:
TERAWAY has developed a first generation of full-photonic THz emitter and receiver modules based on a hybrid photonic integration approach comprising lasers, modulators, optical filters, and photonic mmW/THz emitters and receiver chips [TERAWAY-3]. Special mention should be given to the receiver side, which has been enabled by a novel waveguide-based photoconductive antennas [TERAWAY-4] in order to demonstrate a first-ever photonic integrated circuit (PIC) for full-photonic mmW/THz reception [TERAWAY-5]. As of the transmitter side, photonics-enabled beam steering has been demonstrated using TERAWAY’s mmW/THz emitter arrays hybridly integrated with an optical phased array [TERAWAY-6]. Furthermore, TERAWAY has developed controllers that control the operation of large-scale photonic integrated circuits (PICs) related with the generation, modulation, beamforming and detection of broadband signals in mmW and THz links. Additionally, TERAWAY has showcased its technological concept including optical mmW/THz signal generation and beamforming in a bulk implementation [TERAWAY-7].

5G-Records:
Live Audio use-case: Implementation, integration, and optimization of disaggregated 5G components; optimization of 5G testbed for latency; network integration of live audio production into different 5G testbeds; Proof-of-concept cloud-based remote live audio production with wireless microphones/IEMs; Fully remote-controlled trials and measurements; Proof-of-concept for shared access to spectrum for a private 5G network. Live multi-camera use-case: End-to-end multi-camera live production and remote contribution; Analysis of PTP performances; Design and integration of a 5G portable camera interface unit; Development of the media gateway and the media operational control gateway; validation of the network slicing in the contribution scenario. Testing of a portable 5G standalone setup. FVV (Free Viewpoint Video) use-case: Development of a FVV system; Design and validation of a compact 5G+MEC; Deployment and testing of end-to-end transport slicing over SDN; End-to-end live trial of a music live event; Pioneer tests on immersive content production over millimetre-wave 5G RAN. [5G-RECORDS -1], [5G-RECORDS-2].

DEDICAT 6G:
A first specification of the mechanisms for dynamic coverage and connectivity extension has been released [DEDICAT6G-4]. DEDICAT6G project addresses the design and development of mechanisms for the dynamic coverage and connectivity extension through the exploitation of innovative devices (e.g., drones, robots, connected cars, other mobile assets like forklifts in a warehouse, etc.). The overall aim is to enable the dynamic, opportunistic set up of dynamic coverage and connectivity extensions for covering areas that cannot be easily reached, where infrastructure is required only for a finite, short amount of time, or where regular network infrastructure has been damaged. First release of mechanisms for dynamic distribution of intelligence in the field of architectural techniques has been released [Dedicat6G-5].