2024 EuCNC & 6G Summit - WOS 1570997168

2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

1   Enabling Hexa-X 6G Vision: An End-to-End

2024 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit) | 979-8-3503-4499-8/24/$31.00 ©2024 IEEE | DOI: 10.1109/EuCNC/6GSummit60053.2024.10597113

2  
3  
4  
Architecture
5   Bahare M. Khorsandi1, Mohammad Asif Habibi 2, Giuseppe Avino *3, Sokratis Barmpounakis 4, Giacomo Bernini 5,
6   Mårten Ericson6, Bin Han2, Ignacio Labrador Pavón7, José María Jorquera Valero8, Diego R. Lopez9, Bjoern
7  
Richerzhagen10, Rony Bou Rouphael11, Merve Saimler12, Lucas Scheuvens13, Corina Kim Schindhelm10, Peter
8  
Schneider1, Tommy Svensson14, and Stefan Wunderer1
9   1
10   Nokia, Germany, RPTU, Germany, 3LIST, Luxembourg, 4WINGS, Greece, 5Nextworks, Italy,6Ericsson, Sweden,
2
7
11   Atos, Spain, 8University of Murcia, Spain, 9Telefonica, Spain, 10Siemens, Germany,11Orange, France, 12Ericsson,
12   Turkey, 13TU Dresden, Germany, 14Chalmers, Sweden
13  
14   Abstract*—The end-to-end (E2E) architecture for the 6th description of these UCs can be found in the following and Fig.
generation of mobile network (6G) necessitates a comprehensive 1.
15  
design, considering emerging use cases (UCs), requirements, and
16   key value Indicators (KVIs). These UCs collectively share
17   stringent requirements of extreme connectivity, inclusivity, and
18   flexibility imposed on the architecture and its enablers.
19   Furthermore, the trustworthiness and security of the 6G
20   architecture must be enhanced compared to previous generations,
21   owning to the expected increase in security threats and more
22   complex UCs that may expose new security vulnerabilities.
Additionally, sustainability emerges as a critical design
23   consideration in the 6G architecture. In light of these new set of
24   values and requirements for 6G, this paper aims to describe an
25   architecture proposed within the Hexa-X, the European 6G
26   flagship project, capable of enabling the above-mentioned 6G
27   vision for the 2030s and beyond.
28  
29   Keywords—E2E architecture, sustainability, security, 6G
30  
31   I. INTRODUCTION
Fig. 1: Hexa-X dependability sub-set UCs
32   As the world embraces the full potential of the fifth generation
33   of mobile network (5G), a new era of wireless communication The “Robots to cobots” family prioritizes network flexibility in
34   is already on the horizon. 6G architecture promises to elevate the face of mobility, while also addressing stringent availability
35   connectivity to unprecedented heights and empower a multitude and reliability requirements, coupled with low latencies for
36   of emerging technologies that will redefine the way we direct communication among cooperating mobile robots. Within
37   communicate [1]. At its core, the architecture of 6G is the “Telepresence” family of UCs, the focus lies on the
38   envisioned to be an intricate network of networks, intertwining inclusion of novel Human Machine Interfaces (HMIs) and the
39   satellites, terrestrial infrastructure, and beyond. It will transcend associated demands in terms of bandwidth and low latency. For
40   the traditional boundaries of connectivity, encompassing a wide the “Massive Twinning” family, the potential of a digital
41   range of frequencies, including the sub-terahertz and terahertz representation of the manufacturing environment combined with
42   bands, to facilitate data transfer at much higher speeds. It is a more detailed digital twin of the network environment is being
43   expected that the architecture of 6G will harness innovative studied. This enables the assessment of the impact of human
44   techniques to reduce power consumption and carbon footprint. presence on the next generation of communication system, an
45   As sustainability becomes a global imperative, 6G technology important building block for human-machine collaboration in
will strive to balance the quest for speed and connectivity with flexible manufacturing environments [2].
46  
environmental, societal, and economical responsibility. The rest of this paper is organized as follows: We commence
47  
The design of 6G architecture also requires accommodating the by discussing the requirements of several industrial use cases
48  
requirements of the new UCs with high availability of resources, for 6G in Section II. Then, we highlight key design values that
49  
reliability, and extreme precision. A subset of UCs from Hexa- need to be considered as a design principle for 6G architecture
50   in Section III. Following that, we propose an E2E architecture
51   X [1], focusing on dependability has been selected in this paper
to showcase the technical design reasoning for the proposed for 6G in Section IV. Finally, we conclude the achievements of
52   this paper in Section V.
architecture. Dependability attributes characterizing the
53   application demands for the respective service. A brief
54  
55  
*
56   Giuseppe Avino was part of Telecom Italia (TIM) during preparation of this paper.
57  
60  
61  
62  
63  
64  
65   979-8-3503-4499-8/24/$31.00 ©2024 IEEE 676
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
1
2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

II. REQUIRMENTS OF INDUSTRIAL 6G USE CASES reliability criteria, such applications are especially vulnerable
This section focuses on the main requirements for Hexa-X to any form of network disruptions, no matter how trivial. Often
vision which motivate the technical enablers and the design of entailing critical processes and safeguarding sensitive data, a
the E2E architecture. cybersecurity breach within these realms could precipitate
Inclusion refers to the capability of using and the ease of access substantial losses, jeopardizing life, safety, privacy, and
to a service. This encompasses both technical aspects, such as material assets. Their increasing convergence with human-
coverage (i.e., spatial, and temporal availability of a network cyber-physical systems augments complexity, necessitating a
and service), and user-centric aspects, such as availability and heightened reliance on data and Artificial Intelligence (AI)-
accessibility of the service for a specific group of people, e.g., driven methodologies. This convergence expands the potential
through novel HMIs or interaction with robots. Consequently, attack surface and ushers in fresh vulnerabilities. A more
the term “inclusion” is used interchangeably with “digital detailed exposition of these security concerns can be found in
inclusion.” Inclusion is a key aspect of the “fully merged cyber- [1].
physical worlds” and the “merged reality game/work” UCs, Trustworthiness is intricately linked to both security and
given that they enable participation and interaction in the virtual privacy. While these factors are pivotal, they often remain
(or augmented) world, even if physical barriers (distance, safety opaque to the service user. Building trust necessitates a
of a workplace, disabilities) would hinder those in the real protracted period of consistent, undisturbed service devoid of
world. any security and privacy infringements. To proactively create
Flexibility refers to the ability of the system to adapt to (or trust, black-box solutions must be avoided. Approaches to
undergo adaptation to) changes in its environment and achieving an initial foundation of trust involve presenting
utilization, considering the costs that such changes would security certifications by trusted entities for system equipment,
inflict. This attribute is closely intertwined with the embedding security protocols throughout the supply chain,
dependability and resilience of a system. Flexibility also elucidating insider attack mitigation strategies, and ensuring
impacts the sustainability of the system, as it can allow the adherence to secure operational practices. Defining precise Key
system to address additional (and potentially still unknown) Performance Indicators (KPIs) for security, privacy, and overall
UCs in a resource- and cost-efficient way through re-use and trustworthiness remains a complex endeavour. Preliminary
re-configuration of components. Flexibility also refers to being metrics to consider in this context include:
able to integrate different types of networks (e.g., mesh or Ad- • Strength and quantum-resilience of cryptographic
hoc networks) into the overall network topology. Flexibility is algorithms.
a key requirement in the “interacting and cooperative mobile • Rigor of security policies, exemplified by patterns like re-
robots” and “flexible manufacturing” UCs, as the system must authentication.
adapt to mobility and reconfigurations of the manufacturing • Coverage of security specifications concerning interfaces
environment. and services.
Trustworthiness spans from security to aspects that are • Extent of security assurance coverage.
covered by the dependability framework, such as availability • System's robustness against known attacks targeting
and reliability of a communication service. The overall range of standardized protocols and procedures.
factors contributes to creating a trustworthiness level that Obviously, such KPIs are much vaguer than classical KPIs but
enables a larger set of UCs that wouldn’t be possible with lower still, we consider them useful to get an impression of the
privacy or lower network availability levels. security posture of future 6G standards, technologies, and
Sustainability considers two aspects: sustainability of the 6G architectures.
system itself (e.g., environmental footprint, emissions, and As a concrete approach to address the trustworthiness KVI, we
energy consumption) and the enablement effect 6G systems propose to utilize a Level of Trust Assessment Function
have on different sectors. (LoTAF) that plays a key role in enhancing security by
More in-depth details on trustworthiness and sustainability analyzing the use of applicable security technologies. It serves
aspects are given in Section III. as a neutral and bidirectional service, catering to both trustors
(users), those whom it aids in making informed decisions, and
III. KEY DESIGN VALUES trustees (network providers), those whom it offers insights into
compliance with security requirements and opportunities for
A. Security, privacy and trustworthiness improving service quality.
The 6G network confronts a comprehensive threat landscape. To ensure effectiveness, the LoTAF identifies relevant
Beyond persistent challenges, such as fragile Information and technologies, analyzes risks, and investigates potential attack
Communication Technology (ICT) infrastructure, susceptible patterns. It also leverages AI-driven threat detection
software and cloud configurations, and operational mechanisms and human experience to enhance the
deficiencies, there is the continually evolving cyber-attack identification of cybersecurity threats and attacks. Besides,
paradigm to consider. Distinctly, 6G brings its specific risks, Machine Learning (ML)-based intelligent optimization
stemming from upcoming trends and technologies unique to algorithms enable a balanced configuration of network services,
this generation. These concerns are particularly acute for considering factors like cost and risk based on user-defined
dependability-critical applications. Due to their rigorous criteria.

677
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
2
2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

The LoTAF categorizes the trustworthiness of network services compares 5G NR SA [3] and 6G TCO with two enabler
using a set of thresholds, assigning Level of Trust (LoT) values families, illustrating a notable 26.4% reduction.
to each category. This allows users and providers to make
informed decisions and determine the affinity between users’
security and privacy intents and providers’ capabilities.
Therefore, LoTAF is designed as a service that belongs neither
to network providers nor users, but as a neutral service that
could be deployed, e.g., in a decentralized marketplace where
on-demand resources and services may be provided. In
addition, a continuous update mechanism ensures the
adaptability of the LoT assessments, incorporating feedback
from users and reassessing based on detected attacks. Overall,
the LoTAF constitutes a new and agile way to enhance the
security and privacy of 6G network services, empowering users
and providers with valuable insights and decision-making
support. Fig. 2: TCO comparison between 5G NR SA and 6G for the
"Fully merged cyber-physical worlds" UC.
B. Sustainability
The E2E architecture serves as a foundational element in Societal sustainability: The ICT sector's footprint impacts
addressing the sustainability challenges across economical, include direct effects from infrastructure and devices across
societal, and environmental pillars. In the following we presents their lifecycle, and indirect effects from ICT-enabled solution
the result of our work to address such challenges: deployment. The enablement effect relates to Greenhouse Gas
Economical sustainability: The baseline that we considered in (GHG) emission reductions from implementing an ICT solution
the analysis of Total Cost of Ownership (TCO) is 5G New [4]. Assessing 6G's enablement effect involves comparing it to
Radio Standalone (5G NR SA). A study conducted by the a baseline emission scenario. [6] provides a standard method to
Global System for Mobile Communications Association quantify the enablement effect of existing technologies but
(GSMA) in [3] compares the TCO between 4G and 5G lacks guidance for assessing future ones. This paper offers
considering five key cost items: (i) Radio Access Network comprehensive guidelines for over 30% CO2 emission
(RAN) infrastructure, (ii) energy consumption, (iii) Core reductions in 6G sectors. Various sectors can benefit from
Network (CN) infrastructure, (iv) backhaul, and (v) other solutions like remote industrial control and professional
network-related expenses (e.g., personnel, network consultation. Implementation requires strategic policies for
management). cultural shifts and enhanced collaboration tools. Effective
To compare costs between 5G NR SA and 6G, five key training and fiscal support are essential for maximizing
technical enablers have been identified. These include adoption. Assessing Mixed Reality (MR) UCs and its adoption,
intelligent networks (User Equipment (UE), programmability, challenges persist. Motion sickness, limited mobility with MR
dynamic functions, AI as a Service (AIaaS)), flexible network headsets, and low-resolution visuals are common issues. These
integration (ad-hoc networks, Device to Device (D2D), mesh), problems, often called “cyber-sickness”, result from delays
efficient networks (RAN/CN signaling, function refactoring, between real-world and virtual-world movements [7]. 6G's low
compute as a service), 6G RAN features (high data rate links, latency and edge computing reduce headset weight and enable
Distributed-Multiple Input Multiple Output (D-MIMO), lighter batteries for wireless MR, while also enhancing MR
localization), and service management features focusing on security with robust protocols and encryption for sensitive
continuum management, orchestration, and AI-driven applications.
orchestration. The impact of each enabler on costs depends on Environmental sustainability: While 6G enhances classic
the specific UC. We focus on the “Fully Merged Cyber- KPIs like peak throughput, it strives to reduce energy
Physical Worlds” UC. The analysis quantitatively assesses cost consumption per transported data unit (Wh/bit), or Energy
reductions, focusing on 6G RAN and service management. Our Efficiency (EE), by 90%, in line with past trends. The goal is
initial TCO qualitative analysis indicates that 6G RAN enablers “quasi-zero watt at zero load”, achieved through dynamic
exclusively impact "RAN infrastructure" costs, while service scaling of RAN infrastructure based on real-time traffic load.
management enablers affect "energy consumption" and "other This energy flexibility relies on highly reconfigurable sub-
network-related costs." While 6G RAN enablers have the systems (hardware/software) adaptable to various scenarios.
potential to halve the costs of the RAN infrastructure [4], Assessing EE involves DC power and data volume
allocating the 10% reduction from service management measurements over a specified duration, with ETSI providing a
enablers [5] requires consideration of their baseline weights standardized methodology [8]. However, early 6G design phase
(39% for "energy consumption" and 7% for "other network- EE assessment is complex, requiring model-based evaluations
related costs" [3]), resulting in a 1:5.6 ratio. Consequently, of components. Defining the baseline, such as 5G NR SA with
service management enablers reduce "energy consumption" by high (80%) and low (20%) traffic loads, is crucial. For domain-
8.5% and "other network-related costs" by 1.5%. Fig. 2 specific comparisons, additional details must be considered,
like a 64 TRx setup on FR1 with a 100 MHz bandwidth to

678
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
3
2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

Fig. 3: Hexa-X E2E architecture

analyze a 6G antenna array and RF front-end performance. Key of a network with high heterogeneity of devices, characterized
technology drivers for 6G's tenfold EE improvement include by a wide variety of technologies, in terms of both hardware
but not limited to: (i) RF power amplifiers (ii) Electronic and software. Having extreme-edge as part of the E2E network
components (iii) AI, multi-goals optimization, and adaptive air cloudification can lead to the full integration of cloud-native
interfaces (iv) Sleep modes (v) Path loss (vi) Infrastructure technologies, such as distributed computing and virtualization.
sharing. While these levers focus on the RAN's impact, other The selected key 6G technologies for this layer for further
parts of the 6G system should also contribute to the tenfold EE elaboration are the D-MIMO, Localization & Sensing and NTN.
reduction. D-MIMO enables the use of multiple antennas at both the
transmitter and receiver, leading to improved data transfer rates,
IV. END-TO-END ARCHITECTURE
increased spectral efficiency, coverage extension, and enhanced
Fig. 3 presents an update on the Hexa-X proposed E2E communication reliability. Further on, D-MIMO has the
architecture as well as the technical enablers in a layered potential to allow for further densification of Radio Units (RUs)
structure composed of Infrastructure and cloud layer, Network for increased and consistent area capacity; mitigate shadowing
service layer, and Application layer as well as a cross-layer and blockage thanks to macro diversity; achieve sufficient link
Management and Orchestration (M&O) functionality based on margin despite output power limitations and high pathloss at
[9]. Fig. 3 also demonstrates AI/ML, analytics and security as high frequencies; and allow for lowering Effective Isotropic
features effecting all layers and elements in the architecture. Radiated Power (EIRP) that can simplify deployment. While D-
Following subsections presents each of layers and relative MIMO is not yet supported within 3GPP 5G standards, Multi-
architectural enablers in more details. Transmission Point (multi-TRP) offers a foundation. Real-
A. Infrastrructure and cloud layer world D-MIMO deployment faces challenges in architecture,
Central Processing Units-AP (CPU-AP) split,
This layer plays a crucial role in enabling the seamless fronthaul/backhaul, scalability, and precoding. These issues
communication experience that 6G promises to deliver. It necessitate scalable digital-analog approaches and efficient,
provides physical and virtual resources to host network integrated backhaul/fronthaul solutions. Given the importance
services, cloud applications and application layer [10, 11]. This of densification at high frequencies and available spectrum,
layer is made of a set of technical enablers that work together cost-effective, decentralized solutions are prioritized over
to support the ultra-high speed, reliable, and secure spectral efficiency initially. At low frequencies, the demand for
communication network. At its core, it comprises a network of spectral efficiency favors more centralized, digital strategies.
interconnected devices, including Internet of Things (IoT) and E2E architecture should accommodate diverse D-MIMO nodes,
UE, Base Stations (BSs), small and macro cells, Access Points with a clear separation of control and user planes, enhancing
(APs), cloud infrastructure, etc. The Infrastructure and cloud sustainability by aligning with service demands. Multi-
layer also includes the integration of extreme-edge, that is part connectivity and CPU-AP splits also raise security concerns,

679
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
4
2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

particularly in authentication and MAC layer splitting, necessary AI functionality is included in the network service
suggesting a move towards centralizing user plane security to layer, except the exposure part of AIaaS which belongs to the
enhance network scalability and reduce vulnerabilities. application layer. As shown in Fig. 4 an AIaaS framework can
Localisation and Sensing is one of the new services envisioned be considered and integrated into the 6G architecture to expose
for 6G, residing in the infrastructure layer but also supported by and offer AI capabilities to a wide range of consumers. These
network functionality in the network service layer. One major consumers include M&O, other network functions (belonging
differentiator of 6G compared to previous generations of to different domains, such as RAN and core), application
mobile communication is the vision of integrated sensing and functions, as well as third parties [12]. The objective is to
communication. Based on sensing information, location- and provide a suite of AI functions, including training and
context / environmental-aware services can emerge to form an monitoring, with tailored inference capabilities that depend on
ecosystem of services and applications which may even include the specific consumer's needs. This supports the
network external sensors and localisation systems. The implementation of closed-loop network and service
challenge from an E2E architecture perspective is the automation.
integration of all technical enablers both regarding
communication and sensing, and the management of each
hardware/software component to enable data and information
flow. Well defined interfaces on hardware, protocol and
semantics level allow for new services and applications in the
field of location and environmental context to emerge an
ecosystem of hardware vendors, service, and application
developers and providers to flourish.
Non-Terrestrial Network (NTN) is to provide global coverage Fig. 4: AIaaS framework
and enabling communication in remote and underserved areas. The proposed AIaaS framework comprises 4 functions: AI
As mentioned earlier, an important target of future 6G networks model repository to catalog trained AI/ML models for
is the inclusion. This goal necessitates a flexible NTN deployment. AI training function is dedicated to training ML
architecture for efficient and sustainable 6G. NTNs can models and generating executable versions for integration into
seamlessly integrate with Terrestrial Networks (NTs), AI agents. AI monitoring function assesses the performance of
employing transparent architecture where 5G BS at Ground ML models, triggering retraining when necessary. AI agent
Stations (GS) relay data via satellites directly to UE. The role employs trained ML models for inference and data pre-
of Inter-Satellite Links (ISLs) is pivotal for extending satellite processing. Furthermore, the framework incorporates two
coverage, requiring an architecture that supports efficient signal centralized data stores. One to collect and expose inference
processing for comprehensive global coverage, including data, facilitating the runtime operation of ML models. The
oceans. This involves managing ISLs effectively, especially second stores training data essential for AI training purposes.
given the dynamic nature of satellite movements, which This separation allows for versatility in accommodating diverse
necessitates frequent setup and teardown of links, posing data sources and processing needs.
challenges in 5G/6G networks. To enhance network flexibility The Ad-hoc network is created and controlled by a
and counteract the inherent rigidity of NTNs, mesh ad hoc management network. The primary functionalities that are
network topologies are crucial for the evolution towards 6G. addressed comprise selecting nodes and extreme-edge devices
B. Network service layer for inclusion in the ad hoc network formation, assessing the
trust level of a node for participation in the D2D/mesh network,
This layer is responsible for providing various services to end-
and unifying the modeling of extreme-edge nodes and devices
users [10]. The layer also depicts Network Functions (NFs) and
in terms of their network and computational resources,
their services that are used within the network and are not
capabilities, and constraints. It also includes integrating these
exposed to the end-users. This layer will ensure that users have
elements with network and service orchestration for seamless
access to high-quality, reliable, and secure services [11]. In
management, control, and enforcement, as well as discovering
addition to the communication services, new services such as
nodes and extreme-edge devices, which encompasses
AI and compute, analytics, data collection and localisation and
synchronization aspects for advertising capabilities.
sensing are raising with 6G. NF, operations, and services can
Fig. 5 illustrates the gains in terms of communication resources
be implemented as cloud-native microservices, leading to a
utilization ratio, energy consumption, and cost in a public safety
more softwarised, intelligent, and efficient 6G architecture.
scenario. Timely allocation of excessive communication and
The selected key 6G technologies for this layer for further
computing resource in a remote area is required. A number of
elaboration are AIaaS and Ad-hoc networks.
teleoperated robotic platforms need to navigate and provide
In 6G, AI can enhance in-network enablers such as M&O,
real-time video service for further processing. The scenario
physical layer and resource management, while AIaaS delivers
configuration involves a set of traffic sources, located at various
on-demand AI-related network services, democratizing AI
locations in a remote area (e.g., rural area), a set of candidate
access [11]. 6G offers AI tools as a service, augmenting the
flexible nodes with access and backhauling capabilities
overall network performance (i.e., improving the inclusion)
(Unmanned Aerial Vehicle (UAVs) are assumed in this
and may improve other key values such as sustainability. The

680
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
5
2024 European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit): Wireless, Optical and Satellite
Networks (WOS)

scenario), as well as computing capabilities. There are cryptography, and privacy enhancing technologies apply to
predefined cost vectors associated to each configuration and functions and procedures on all layers. Security and risk
network node. The results compare the KPIs in the case of a management must evolve towards AI/ML-driven automated
static network infrastructure, i.e., a macro cell-based closed loop mechanisms. Distributed ledger technologies may
deployment (blue plot), against the flexible allocation of a set facilitate distribution of trust in interdomain management. To
of UAV-powered network formations (green plot). According ensure system integrity at boot and runtime, attestation
to the results, certain trade-offs need to be considered, technologies must be applied, building on trust foundations,
depending on the amount of resources that need to be allocated; such as trusted execution environments, in the secure
for example, for very high number of UAVs the cost may infrastructure and cloud layer. Further in this layer, mechanisms
exceed the static infrastructure costs; nevertheless, there are must be applied to mitigate the inherent threat of jamming
clear gains in terms of energy consumption when attacks. Moreover, physical layer security technologies and
commissioning only the required number of flexible network quantum key distribution may complement the cryptographic
nodes. schemes used today.
V. CONCLUSION
This paper provided the Hexa-X vision on 6G E2E architecture.
It developed based on 6G core values namely sustainability,
inclusion, and trustworthiness in order to satisfy the UC
requirements. The design of the 6G architecture is flexible and
highly specialized to accommodate a comprehensive and highly
advanced wireless ecosystem. The main enablers for improved
flexibility and inclusions described in this paper are D-MIMO,
NTN, and ad-hoc networks. The sensing and AIaaS enablers
extend the mobile network for new 6G UCs and, finally, the
M&O enabler using the cloud continuum improves flexibility
and trustworthiness.
ACKNOWLEDGEMENTS
Fig. 5: Comparison of static infrastructure with flexible UAV-
based nodes in a disaster scenario. This work was supported by the European Commission
through the H2020 Project Hexa-X under Grant 101015956.
C. Application layer
REFERENCES
The topmost layer is Application layer which interacts directly
with end-user applications, facilitating the exchange of data and [1] Hexa-X, “Deliverable D1.4: Hexa-X architecture for B5G/6G networks –
final release”, July 2023.
information [12].
[2] Hexa-X, “Deliverable D7.3: Special-purpose functionalities: final
D. Management and Orchestration (M&O) solutions”, May 2023.
[3] GSMA, “5G-era Mobile Network Cost Evolution,” August 2019.
The M&O functionality, depicted as a vertical blue rectangle in
[4] H. Viswanathan, et. al., "Energy efficiency in next generation mobile
the architecture, spans three layers to orchestrate resources at networks," November 2022.
the infrastructure level and manage services at the network [5] H. Chergui, et. al., "Zero-Touch AI-Driven Distributed Management for
service layer for application layer services. It aims to enhance Energy-Efficient 6G Massive Network Slicing," in IEEE Network, vol.
flexibility and trust through a "device-edge-cloud" continuum, 35, no. 6, pp. 43-49, November/December 2021.
extending 5G network slicing to include all network resources [6] ITU-T L.1400 - Overview and general principles of methodologies for
assessing the environmental impact of information and communication
and even those beyond the MNO's (Mobile Network Operator) technologies, 2023.
scope [10]. This approach integrates extreme-edge and [7] I. Akyildiz, et. al., “Wireless communication research challenges for
addresses the volatility of device connections. To support Extended Reality (XR).” ITU Journal on Future and Evolving
service and resource orchestration, M&O combines core Technologies (ITU J-FET), 2022.
capabilities with AI/ML for predictive actions, security across [8] ETSI Standard, “Environmental Engineering (EE); Assessment of mobile
levels, and advanced monitoring for comprehensive analytics, network energy efficiency”, ETSI ES 203 228 V1.3.1 (2020-10).
thereby facilitating orchestration decisions. [9] Khorsandi, Bahare M., et al., "6G E2E Architecture Framework with
Sustainability and Security Considerations." IEEE Globecom Workshops
E. Security Architecture (GC Wkshps), 2022.
[10] Hexa-X, “Deliverable D6.2: Design of service management and
The 6G E2E architecture must be accompanied by a security orchestration functionalities”, April 2022.
architecture specifying what security and privacy mechanisms [11] Hexa-X, “Deliverable D5.3: Final 6G architectural enablers and
must be applied, and how they interact to ensure overall technological solutions”, April 2023.
security, privacy and trustworthiness. Essential 6G security [12] M. A. Habibi et al., "Toward an Open, Intelligent, and End-to-End
technology enablers and their placement in the architecture is Architectural Framework for Network Slicing in 6G Communication
Systems," in IEEE Open Journal of the Communications Society, vol. 4,
highlighted in Fig. 3. Secure AI/ML, quantum-safe pp. 1615-1658, 2023.

681
Authorized licensed use limited to: Politecnico di Torino. Downloaded on July 29,2024 at 16:05:25 UTC from IEEE Xplore. Restrictions apply.
6