Content uploaded by Madhusanka Liyanage
Author content
All content in this area was uploaded by Madhusanka Liyanage on Mar 30, 2024
Content may be subject to copyright.
Advancing Security for 6G Smart Networks and
Services
Madhusanka Liyanage∗, Pawani Porambage†Engin Zeydan‡, Thulitha Senavirathne§,
Yushan Siriwardhane¶, Awaneesh Kumar Yadav∥, Bartlomiej Siniarski∗∗
∗§ ¶∥School of Computer Science, University College Dublin, Ireland, †VTT Technical Research Centre, Finland
‡Centre Tecnol`
ogic de Telecomunicacions de Catalunya (CTTC), Spain, ¶CWC, University of Oulu, Finland
Email: ∗madhusanka@ucd.ie, †pawani.porambage@vtt.fi, ‡engin.zeydan@cttc.cat,
§thulitha.senevirathna@ucdconnect.ie, ¶yushan.siriwardhane@oulu.fi, ∥∗∗bartlomiej.siniarski@ucd.ie
Abstract—6G Smart Networks and Services are poised to
shape civilization’s development of 2030’s world, supporting the
convergence of digital and physical worlds. The arrival of 6G
networks brings unprecedented challenges and opportunities,
requiring robust security measures to safeguard against emerging
threats. Thus, several complementary issues must be addressed
to advance the security of 6G smart networks and services.
This research paper explores a multi-faceted approach to 6G
security, addressing key areas of security of 6G smart networks
and services such as distributed trusted AI/ML, zero-touch
holistic end-to-end (E2E) security, energy efficient security and
privacy enablers, real-time resilience for timing-sensitive 6G
software technologies and quantum-safe 6G communications. We
comprehensively investigate the security and privacy challenges
associated with integrating these technologies into 6G networks
and their possible direction to mitigate them.
I. INTRODUCTION
The evolution of communication technologies to 6G in-
troduces new dimensions of connectivity and capabilities.
However, with increased capabilities and advanced services
comes the need for robust security measures. This research
aims to propose a comprehensive security framework for
6G networks, addressing various challenges and leveraging
advanced technologies.
6G networks, shaping civilization into the 2030s, will
merge digital and physical worlds, enabling advanced ap-
plications like the metaverse and immersive communication.
These networks will feature joint sensing, programmability,
energy efficiency, trustworthiness, scalability, and affordability.
However, their complexity requires a new approach to network
management, and their advanced capabilities necessitate robust
security against the expanding threat landscape. The strategic
integration of AI/ML is pivotal for every layer of upcoming
6G networks, with applications in automated decision-making,
enhancing network performance with intelligence decisions,
ensuring zero-touch automation, and automated end-to-end
security rely on AI integration in 6G. To enable such ubiq-
uitous AI usage in 6G demands efficient data management
solutions for collecting, processing, storing, and exposing
information. Balancing the power of AI, laden with potentially
private data, underscores the critical need for privacy and
safety assurance. Quantifying the security and privacy impacts
throughout the AI pipeline is a significant challenge. Thus,
several complementary issues should be addressed to advance
the security of 6G smart networks and services:
Exploitation of (distributed) trusted AI/ML for 6G infras-
tructures: 6G networks must secure the entire AI life cycle for
predictable models and behaviours, addressing vulnerabilities
during development and deployment, considering accountable
AI measures, and implement protection mechanisms against
potential misuse.
Zero-touch Integrated E2E Security Deployment for Adap-
tive Security: Securing 6G involves innovative approaches
like Zero-touch and zero-trust frameworks while ensuring
compliance with human-centric measures. These mechanisms
should support E2E security by supporting a complete cycle
of digital infrastructure resilience, i.e. threat identification,
protection, detection, response, and recovery.
Innovative Enablers for Energy Efficient Security and
Privacy : Enabling energy-efficient security and privacy in
6G could employ various techniques such as energy-efficient
ML/AL architectures, multi-level security, spatial fragmen-
tation, flexible profiling of resources and energy-efficient
privacy-enhancing technologies. To promote green and sus-
tainable AI methodologies by achieving energy efficiency in
6G network security mechanism design.
Real-Time Resilience using Adaptive Time Sensitive Soft-
ware Technologies for 6G Service Provisions: 6G de-
ployment involves timing-sensitive software and hardware
for multi-stakeholder service provision. Designing adaptive
security and privacy mechanisms for these Time-Sensitive
Networks (TSNs) is important to support future 6G smart
networks and services.
Quantum-Safe 6G Communications Securing 6G commu-
nications should be targeting on integrating Quantum key
distribution and post-quantum cryptography for long-term net-
work security, including challenges that could arise in a post-
quantum era.
In the remaining sections, this paper discusses the above-
mentioned prioritizing trusted AI, zero-touch security, energy-
efficient measures, quantum-safe communications, and real-
time resilience, which are vital for robust 6G network security.
II. DISTRIBUTED AND TRU ST ED AI/ML F OR 6G
The use of AI and machine learning in 6G networks is
pivotal but raises concerns about security. This section focuses
on developing frameworks that ensure the security of the entire
AI lifecycle. This involves addressing vulnerabilities in AI
Adversaries
Learn using AI
Adversaries
Attack 6G/AI
Models
AI in 6G
Operation
Adversaries
using AI
AI in 6G
Attacked
Resilient
AI in 6G
AI in 6G
Security
- Attacks on Learning Phase
- Attacks on Inference Phase
- Model's API based Attacks
- Detect Vulnerabilities
- Predict next Attacks
Stolen Data,
Learned AI Models
Poisoning Data
Extracted AI Models
Adversaries
Cloud AI
1 2
Centralized Cloud
6G Core
Automated Network and
Service Management Layer
Intelligent
Services
Cross-domain
Connectivity
1 2
1
2
3
4
5
AI in 6G as an Enabler
AI in 6G as a Defender
AI as an Offender
AI in 6G as a Target
Resilient AI in 6G
1
2
Edge
AI
6G Tiny
Cell
AI based Security Deployed
- Authentication/Authorization
- Anomaly Detection with AI
Compromised
V2V
Resilient AI Systems Deployed
- Moving Target Defense
- Adversarial ML
- Input Validation
- Differential Privacy
4
Edge AI Attacked
- Poisoning Attacks
- Model Evasion
- Model Extraction
- Model Inversion
5
Transport/
Delivery
Intelligent Transportation/ V2V
Extended Reality (XR) Industry 5.0
Brain Computer Interactions
XAI Security in
6G Core
XAI Security Network
and Service
Management
XAI Security in
Application Layer
Benign
6G Applications
Fully Distributed
Learning Network
of Autonomous
Cars
X
A
I
Fully Distributed
AI for Drones
Fig. 1: Different roles of AI in 6G smart networks.
models, quantifying model vulnerability, and implementing
protective measures against AI misuse.
A. AI’s role as an Enabler, Defender, Offender, and a Target
in 6G Smart Networks and Services
Fig. 1 depicts the application of distributed AI techniques in
6G networks and applications while highlighting the various
roles AI plays in 6G service provisioning. The network-wide
distributed deployment of AI techniques is 6G rather than
confining it to server environments. In the 6G era, networks
and applications utilize AI as an enabler for efficient service
provision due to AI’s ability to learn in complex environments
and make accurate predictions. AI plays a vital role in 6G
network security provision as a defender. The interconnection
of complex networks in 6G that support applications with
diverse requirements makes the security provision challenging.
Proactive approaches where the network can predict an attack
based on the behavior of traffic would be required instead
of conventional reactive approaches. AI’s ability to uncover
anomalies within large volumes of data can prevent the net-
work and applications from zero-day attacks.
B. Federated AI/ML
Not only the network and application designers but also the
adversaries get access to AI. This defines the AI’s role as an
offender to the network where the adversaries utilize AI as
a tool to learn about the network and launch intelligent and
adaptive attacks towards the 6G system. The adversaries can
collect data about the network vulnerabilities using AI models
and predict the most suitable points to attack for maximum
impact. The distributed AI model deployed in 6G can also be a
target for adversaries. In this case, the adversaries aim to attack
the AI models and modify the output produced by the models
towards an attacker’s objective. Data and model poisoning
attacks, inference attacks, model inversion, and model evasion
are some of the well-known attacks against AI systems [1].
Hence the AI systems should be designed with enhanced
robustness and resilience against such attacks so that the true
potential of AI can benefit 6G service provision.
Adapting the distributed paradigm of AI/ML in the realms
of 6G takes the sole accountability off from the monolith of
centralized servers and disseminate it across several devices.
The centralized servers which solely controlled the learning
process will now have a number of members that would have
a say in the final model.
Federated AI/ML in 6G can span across several layers of
the 6G architecture such as edge layer, user equipment (UE)
layer and application layers. Edge computing the necessary
infrastructure for the 6G decentralized AI/ML applications.
This architecture for intelligent applications, commonly known
as Edge AI, is aimed to improve the efficiency of bandwidth
usage and reduce the latency of communication in the high de-
manding era of 6G. On the other hand, current AI/ML methods
in 5G requires large data storage spaces and computational
resources to cater the services of a largely evolving hetero-
geneous UEs. Apart from the advantage of optimal usage of
computing power, federated AI can also provide a layer of
privacy to the users. Even though with secure transmissions
and regulatory requirements, the third-party AI/ML service
providers could still be subjected to privacy leakages either
from third party attackers or the service providers themselves.
Pertaining to proprietary rights, certain organizations are not
too interested in sharing their data with the cloud service
providers after all. Also, sending data to cloud platforms
for processing is an added cost to the organizations making
edge/UEs viable in the business perspective.
C. Trust and accountability
In federated AI/ML, regardless of architectural differences,
trust is crucial, particularly in applications like automated
vehicle networking [2] and edge communication [3]. While
FL enhances privacy, its expanded threat landscape due to
device connectivity necessitates transparent decision-making
for trustworthiness. Explainable Artificial Intelligence (XAI)
is vital for security in Beyond 5G networks and is integral
to FL applications’ trust. XAI methods, including correlation
analysis and data visualization, provide essential insights into
training data, influencing AI/ML model development. This is
particularly relevant for FL’s diverse data types and structures.
Aggregation in federated learning (FL) usually occurs in
cloud or edge environments, with security depending on
whether it’s a Trusted Execution Environment. In trusted
environments, the aggregator is generally secure, but in un-
trusted ones, FL faces risks from client-originated attacks like
data poisoning. Detecting and isolating malicious elements
in FL is feasible through XAI methods, such as identifying
poisoned data in health devices or spotting abnormal data
points. Continuous monitoring of the aggregator using XAI,
especially in untrusted environments, is crucial. Techniques
like SHAP are effective for detecting shifts in ML models,
helping to counter evasion attacks [4], a significant threat in
the Beyond 5G era.
III. ZERO -TO UC H INT EG RATE D E2E SECURITY
DEP LOY ME NT F OR ADA PT IV E SECURITY
The integration of AI, quantum technologies and end-to-
end security solutions provides a coherent framework for
establishing an adaptable zero-touch security implementation
that meets the unique challenges of highly distributed 6G
environments [5]. As illustrated in Fig. 2, there are many
challenges in the context of highly distributed and virtualized
6G environments such as quantum-resistant cryptography, pri-
vacy concerns, user consent and control, resilience to advanced
threats, AI model security, interoperability and standards. The
design and implementation of adaptive ciphers, post-quantum
cryptographic algorithms, robust model validation and test-
ing protocols, data anonymization and encryption techniques,
mechanisms to isolate and secure third-party applications or
attack surface limitations can play a crucial role in the zero-
trust integration of adaptive E2E security.
A. Enhanced Security Mechanisms in 6G Environments
The distributed and virtualized nature of 6G environments
introduces new challenges in securing the network. Strategies
such as network segmentation, microsegmentation, and dy-
namic network policies are essential in minimizing the attack
surface for network orchestration. AI-driven anomaly detec-
tion plays a crucial role in continuously monitoring network
activity to identify and isolate potential threats. The integra-
tion of quantum-resistant security protocols into zero-touch
security orchestrators is vital for reducing vulnerabilities by
strengthening communication channels against quantum-based
attacks [6]–[8]. Additionally, with the increasing reliance
on third-party applications in 6G ecosystems, implementing
robust mechanisms to isolate and secure these applications is
imperative. Quantum-safe algorithms are crucial in protecting
against quantum threats. Technologies like containerization
(e.g., Docker, Kubernetes), hardware-based isolation mecha-
nisms (e.g., Intel SGX, AMD SEV), and Trusted Execution
Environments (e.g., Intel SGX, ARM TrustZone) create secure
Fig. 2: The orchestration approach based on Zero Trust inte-
grated into a mobile network infrastructure and corresponding
security and privacy challenges
execution environments. These are complemented by dynamic
sandboxing and AI-powered behavioral analytics for granular
control over applications, and blockchain-based application
isolation with a Zero Trust Network Access approach for
managing permissions and continuously verifying identities.
B. Adaptive Security and Resource Optimization in 6G
An essential aspect of securing 6G networks involves the de-
velopment and implementation of adaptive ciphers [9]. These
ciphers dynamically adapt encryption algorithms and key
management protocols in response to the evolving threat land-
scape. Utilizing AI algorithms, adaptive ciphers continuously
evaluate network conditions, traffic patterns, and potential
vulnerabilities to optimize encryption strategies in real-time.
This proactive approach is critical for maintaining a robust
defense against sophisticated cyberattacks. Furthermore, AI
algorithms are central in the dynamic allocation of security
resources, ensuring optimal protection without compromising
network performance. Quantum-based key distribution mech-
anisms [10] contribute to resource efficiency by providing
secure communication channels with reduced computational
overhead. Efficient orchestrator design is key to balancing
security measures with available resources, optimizing the
trade-off between security and operational efficiency in 6G
networks.
IV. REA L- TIME RESILIENCE USING ADAPTIVE TIME
SENSITIVE SOF TWAR E TECHNOLOGIES FOR 6G
Time synchronization is crucial for the efficient operation
of deterministic communication systems, ensuring seamless
integration between the physical and digital worlds. For ex-
ample, autonomous vehicles rely on exact time alignment to
operate safely and efficiently. This necessity extends to all
network notes, which require access to a shared time reference
to support various time-sensitive applications. Firstly, it’s es-
sential for network nodes to synchronize with one another with
high accuracy and precision. This synchronization enables the
execution of time-triggered operations in line with a globally
coordinated schedule. Secondly, applications at the endpoints
must also be time-aware, undertaking specific actions at pre-
determined moments. In the context of 6G networks, the
importance of time synchronization is amplified due to the
expected increase in network speed and capacity.
A. Challenges and Solutions for Time-Synchronization in 6G
With 6G’s potential to support more complex and time-
critical applications, such as advanced autonomous systems
and IoT devices, maintaining precise time alignment becomes
even more critical. 6G technology is expected to facilitate
ultra-low-latency interactions crucial for critical applications,
notably in industrial automation. This advancement necessi-
tates the implementation of highly accurate and swift time
synchronization mechanisms. Essential to this process is the
incorporation of end-to-end (E2E) security measures, which
rely on precise monitoring and rapid response capabilities
to effectively implement countermeasures against potential
threats. A challenge in this setup is that network monitoring,
while essential for security, tends to introduce additional
latency. It is crucial that this added latency remains minimal to
ensure it does not compromise the stringent timing guarantees
required for effective communication in these applications.
The future of E2E deterministic networks is likely to continue
leveraging standards like TSN or Deterministic Networking
(DetNet), traditionally used in secure, controlled environments
to mitigate exposure to external threats. Due to the strict
timing requirements of TSN/DetNet, potential vulnerability
points will be introduced in an integrated 6G-TSN network.
For instance, the need for accurate time synchronization in a
TSN network makes DoS attacks very effective, e.g., a time-
critical flow can be disrupted by a long PTP outage, slowDoS-
type attacks can be sufficient to disrupt sensitive applications
[11]. These attacks are difficult to be detected with traditional
signature-based intrusion detection approaches, even more
when the HTTP traffic is encrypted. However, the landscape
for 6G systems is significantly different, characterized by open
and heterogeneous networking technologies spanning various
sectors. This openness increases the vulnerability to external
attacks, such as jamming, device impersonation, and the intro-
duction of compromised or fake nodes. Such threats can have
immediate and significant impacts on time synchronization,
determinism, packet delivery, and overall network consistency.
B. Future landscape and security enablers
To effectively address these challenges, a comprehensive
’security by design’ strategy is essential. This strategy includes
implementing Zero-touch network and Service Management
(ZSM) for an automated, end-to-end security management pro-
cess. Through the integration of ZSM, 6G networks are posi-
tioned to significantly improve their resilience and adaptability.
This integration is crucial for meeting the high standards of
performance and reliability expected of these networks, while
also upholding robust security protocols in a complex and
vulnerable digital environment. This approach is critical in
meeting the requirements of Cyber-Physical Systems (CPS)
applications and support dynamic network changes.
Furthermore, 6G systems are set to adopt even greater
adaptability in terms of application operation modes. This
means applications will have the capacity to adjust their op-
erational modes dynamically, aligning with real-time network
conditions and specific application needs. Such flexibility is
key to promptly adapting to fluctuating scenarios, thereby
ensuring consistent and robust support for a range of CPS
applications. This paradigm shift towards enabling determin-
istic communications within 6G systems hinges on a variety
of technological advancements across different sectors of the
communications and computing infrastructure.
Next-generation networks are addressing security challenges
in time-sensitive communication with a multi-faceted ap-
proach. Machine learning algorithms will analyze system
behavior for vulnerability prediction and network performance
optimization. Digital twinning, creating virtual replicas of net-
work components, allows for controlled behavior simulation.
The integration of TSN and DetNet with wireless technology
enhances reliability and flexibility for mobile applications
in 6G networks. Emphasizing ’Security by Design’, security
features are integrated at the initial stages of network devel-
opment, making security a fundamental part of the network
architecture, essential for addressing sophisticated challenges
in future networks.
Standards such as IEEE 802.1Q [12] (foundational TSN ele-
ments), 802.1Qbv [13] (TSN traffic scheduling), 802.1Qci [14]
(filtering and plicing), 802.1CB [15] (reliability improvements)
collectively form a comprehensive framework for implement-
ing and managing TSN in various network environments,
including the upcoming 6G networks, where the demand for
low-latency and high-reliability communication is expected to
increase significantly.
V. IN NOVATIV E ENABLERS FOR ENERGY EFFICIENT
SECURITY AND PRI VACY
The focus on energy efficiency implies a concerted ef-
fort to minimize energy consumption or optimize resource
usage within the 6G network infrastructure. This is crucial
for sustainability and reducing the environmental impact of
network operations. This is particularly important as networks
become increasingly complex and resource-intensive and more
emphasis is placed on the development of sustainable and eco-
friendly communication and network technologies [16]. As
summarized in Figure 3, this section explores technologies
such as multi-level security, security segregation, and privacy
quantization. It also shows how to enable efficient security and
privacy measures, including multi-stakeholder moving target
defense developments and rapid proactive security recovery
techniques.
Fig. 3: Energy efficient security and privacy for 6G.
Multi-level security in a telecommunications network refers
to the implementation of security measures at different levels
within the network infrastructure to protect against different
types and levels of security threats. The concept is particularly
relevant in environments such as 6G networks, where sensitive
information, critical services and different stakeholders coex-
ist. Implementing multi-level security with a focus on energy
efficiency involves combining security measures at different
levels of the network architecture while optimizing energy
consumption [17]. Key aspects of multi-level security in 6G
networks may include various security measures with adaptive
authentication and authorization and also selective data en-
cryption techniques; the use of sustainable AI/ML algorithms
for security operations; the adoption of energy-aware security
policies; and compliance with industry regulations; distributed
security measures with dynamic security adjustments; energy-
efficient security devices or energy harvesting for security
devices; zero-trust security models with fine-grained access
control; hierarchical security architectures and security levels
that have a network of networks with edge-level security
processing.
In the context of 6G networks, security segregation refers
to the practice of dividing and isolating different components,
data sources, applications, stakeholders, functions or entities
within the network architecture to enhance security. The aim
is to increase security by creating boundaries and controlling
interactions between different entities within the 6G echo
system. Energy-efficient security segregation in 6G networks
involves implementing measures to separate and protect dif-
ferent levels of network resources and data while optimizing
energy consumption. This approach aims to improve security
without compromising the overall energy efficiency of the net-
work, e.g. through dynamic and context-aware segmentation
and resource-aware security policies. Novel techniques such as
privacy quantization have also attracted a lot of attention in the
6G research community to improve privacy while increasing
energy efficiency. In the application of differential privacy in
federated learning, for example, the use of quantization proves
to be a key technique for reducing the information encoded
over the original input. An interesting question is to explore
whether the introduction of randomization in quantization
schemes can effectively optimize the trade-off between privacy
preservation and model accuracy in the context of differentially
private federated learning [18].
The focus of the above-mentioned aspects of energy effi-
ciency lies in enabling effective security and privacy measures
in the 6G context. In some ways, they can be specifically
addressed with the implementation of multi-stakeholder mov-
ing target defense evolutions, which are dynamic security ap-
proaches that adapt to evolving threats by frequently changing
the attack surface [19]. In addition, work can be extended
to the deployment of rapid, proactive security recovery tech-
niques aimed at quickly detecting and addressing security
incidents. Incorporating these strategies should improve the
overall resilience and responsiveness of security mechanisms
in the 6G network environment and ensure energy efficiency.
VI. QUA NT UM -SAF E 6G COMMUNICATIONS
The proliferation of Internet of Things (IoT) devices is
expected to increase significantly due to the services offered
by 6G. The security and privacy of the 6G core network
will be closely linked to the security of IoT devices and
their protection mechanisms [20]. In the current landscape,
researchers predominantly use either symmetric or asymmetric
encryption approaches to secure communications. The security
of asymmetric encryption is based on the assumption of the
hardness of factorization and discrete logarithm problems.
However, Shor [21] has shown that post-quantum computers
have the ability to solve these problems efficiently. Conse-
quently, existing security mechanisms based on factorization
and discrete logarithm problems, such as RSA and ECC,
may be insufficient for ensuring communication security.
Furthermore, Grove’s findings indicate that small key sizes
in existing symmetric algorithms such as SNOW 3G, AES
and ZUC are also vulnerable to quantum attacks [22]. In
the 5G specification, symmetric algorithms such as SNOW
3G, AES and ZUC use 128-bit keys for both encryption and
message integrity. In the post-quantum world, however, these
key lengths are considered insufficient due to the Grover algo-
rithm, which requires doubling the key lengths of symmetric
primitives. Therefore, symmetric cryptography with at least
256-bit keys must be implemented for future 6G networks in
order to maintain the current level of security against quantum
attacks. Fortunately, block ciphers such as AES remain secure
and applicable in this context. The same applies to modern
cryptographic hash functions such as SHA-2 and SHA-3.
This revelation is a crucial signal for security researchers to
start exploring alternative cryptographic solutions, considering
the potential impact of quantum computing on the security
landscape for 6G communications [23].
Different types of post-quantum cryptography algorithms
and approaches have so far been developed to counter the
threat posed by quantum computers to classical encryption
methods. These post-quantum cryptographic methods can be
categorized into different families based on their mathematical
foundations and techniques. Here are some of the major types
of post-quantum cryptography such as lattice-based cryptogra-
phy, code-based cryptography, multivariate cryptography, and
hash-based cryptography. Figure. 4 discusses the components
Fig. 4: Quantum-Safe 6G Communications
of quantum-safe 6G communications. On the other hand, there
are currently no secure post-quantum cryptographic algorithms
that can provide both very small keys and compact cipher-
texts/signatures while ensuring efficient key generation, en-
cryption and decryption or signing and verification processes.
In the transition from today’s asymmetric primitives to safe
post-quantum alternatives, trade-offs are unavoidable. Such
replacements inevitably involve costs that affect either the
communication or the operational efficiency of the network.
It is imperative to conduct further research to determine
the optimal application of secure post-quantum cryptography
and to ensure compliance with the envisioned performance
and functionality of the 6G architecture. Striking the right
balance between security and operational efficiency will be
critical when it comes to adapting cryptographic solutions to
the evolving landscape posed by the challenges of quantum
computing for 6G communications.
VII. CONCLUSION
In conclusion, this research paper proposes a comprehensive
framework for 6G security that considers various dimensions,
including AI/ML for intelligent threat detection and mitigation,
E2E security to address vulnerabilities across the entire com-
munication network, efficient enablers to optimize resource
allocation and usage, zero-touch deployment for seamless and
secure network setup, quantum technologies for advanced
encryption methods, time-sensitive software to ensure real-
time responsiveness, and privacy considerations to safeguard
user data. The detailed integration of these elements aims
to create not just a robust but also a resilient and provable
security foundation for the next generation of communication
networks. By exploring these key areas and addressing the
intricate interplay between them, we pave the way for a secure
and resilient foundation for the future of 6G networks.
ACK NOW LE DG EM EN T
This work is partly supported by the European Union under
the Robust-6G project (Grant ID. 101139068) and by Science
Foundation Ireland under CONNECT phase 2 (Grant no.
13/RC/2077 P2) projects.
REFERENCES
[1] Y. Siriwardhana, P. Porambage, M. Liyanage, and M. Ylianttila, “AI
and 6G Security: Opportunities and Challenges,” in 2021 Joint Eu-
ropean Conference on Networks and Communications & 6G Summit
(EuCNC/6G Summit), 2021, pp. 616–621.
[2] A. Renda, P. Ducange, F. Marcelloni, D. Sabella, M. C. Filippou,
G. Nardini, G. Stea, A. Virdis, D. Micheli, D. Rapone et al., “Federated
learning of explainable ai models in 6g systems: Towards secure and
automated vehicle networking,” Information, vol. 13, no. 8, p. 395, 2022.
[3] W. Y. B. Lim, N. C. Luong, D. T. Hoang, Y. Jiao, Y.-C. Liang, Q. Yang,
D. Niyato, and C. Miao, “Federated learning in mobile edge networks:
A comprehensive survey,” IEEE Communications Surveys & Tutorials,
vol. 22, no. 3, pp. 2031–2063, 2020.
[4] G. Fidel, R. Bitton, and A. Shabtai, “When explainability meets adver-
sarial learning: Detecting adversarial examples using shap signatures,” in
2020 international joint conference on neural networks (IJCNN). IEEE,
2020, pp. 1–8.
[5] A. Hummelholm, T. H¨
am¨
al¨
ainen, and R. Savola, “Ai-based quantum-
safe cybersecurity automation and orchestration for edge intelligence in
future networks,” in Proceedings of the European Conference on Cyber
Warfare and Security. Academic Conferences International, 2023.
[6] E. Zeydan, J. Baranda, and J. Mangues-Bafalluy, “Post-quantum
blockchain-based secure service orchestration in multi-cloud networks,”
IEEE Access, vol. 10, pp. 129 520–129 530, 2022.
[7] L. Malina, P. Dobias, J. Hajny, and K.-K. R. Choo, “On deploying
quantum-resistant cybersecurity in intelligent infrastructures,” in Pro-
ceedings of the 18th International Conference on Availability, Reliability
and Security, 2023, pp. 1–10.
[8] E. Zeydan, Y. Turk, B. Aksoy, and Y. Y. Tasbag, “Post-quantum era in
v2x security: Convergence of orchestration and parallel computation,”
IEEE Communications Standards Magazine, vol. 6, no. 1, pp. 76–82,
2022.
[9] S.-C. Lin, K.-C. Chen, and A. Karimoddini, “Sdvec: Software-defined
vehicular edge computing with ultra-low latency,” IEEE Communica-
tions Magazine, vol. 59, no. 12, pp. 66–72, 2021.
[10] M. Mehic, M. Niemiec, S. Rass, J. Ma, M. Peev, A. Aguado, V. Martin,
S. Schauer, A. Poppe, C. Pacher et al., “Quantum key distribution:
a networking perspective,” ACM Computing Surveys (CSUR), vol. 53,
no. 5, pp. 1–41, 2020.
[11] N. Garcia, T. Alcaniz, A. Gonz´
alez-Vidal, J. B. Bernabe, D. Rivera,
and A. Skarmeta, “Distributed real-time slowdos attacks detection over
encrypted traffic using artificial intelligence,” Journal of Network and
Computer Applications, vol. 173, p. 102871, 2021.
[12] “Ieee standard for local and metropolitan area networks–bridges and
bridged networks,” IEEE Std 802.1Q-2022 (Revision of IEEE Std
802.1Q-2018), 2022.
[13] “Ieee standard for local and metropolitan area networks – bridges and
bridged networks - amendment 25: Enhancements for scheduled traffic,”
IEEE Std 802.1Qbv-2015, pp. 1–57, 2016.
[14] “Ieee standard for local and metropolitan area networks–bridges and
bridged networks–amendment 28: Per-stream filtering and policing,”
IEEE Std 802.1Qci-2017, pp. 1–65, 2017.
[15] “Ieee standard for local and metropolitan area networks–frame replica-
tion and elimination for reliability,” IEEE Std 802.1CB-2017, 2017.
[16] P. Porambage and M. Liyanage, Security and Privacy Vision in 6G: A
Comprehensive Guide. John Wiley & Sons, 2023.
[17] P. Porambage, J. Pinola, Y. Rumesh, C. Tao, and J. Huusko, “Xcaret:
Xai based green security architecture for resilient open radio access
networks in 6g,” in 2023 Joint European Conference on Networks and
Communications & 6G Summit (EuCNC/6G Summit). IEEE, 2023.
[18] N. Lang, E. Sofer, T. Shaked, and N. Shlezinger, “Joint privacy en-
hancement and quantization in federated learning,” IEEE Transactions
on Signal Processing, vol. 71, pp. 295–310, 2023.
[19] T. Zhang, C. Xu, P. Zou, H. Tian, X. Kuang, S. Yang, L. Zhong, and
D. Niyato, “How to mitigate ddos intelligently in sd-iov: a moving target
defense approach,” IEEE Transactions on Industrial Informatics, vol. 19,
no. 1, pp. 1097–1106, 2022.
[20] J. Partala, “Post-quantum cryptography in 6g,” 6G Mobile Wireless
Networks, pp. 431–448, 2021.
[21] P. W. Shor, “Polynomial-time algorithms for prime factorization and dis-
crete logarithms on a quantum computer,” SIAM Journal on Computing,
vol. 26, no. 5, pp. 1484–1509, 1997.
[22] L. K. Grover, “A fast quantum mechanical algorithm for database
search,” in Proceedings of the twenty-eighth annual ACM symposium
on Theory of computing, 1996, pp. 212–219.
[23] G. G. Rozenman, N. K. Kundu, R. Liu, L. Zhang, A. Maslennikov,
Y. Reches, and H. Y. Youm, “The quantum internet: A synergy of
quantum information technologies and 6g networks,” IET Quantum
Communication, vol. 4, no. 4, pp. 147–166, 2023.
