A Security and Trust Framework for
Decentralized 5G Marketplaces
Jos´
e Mar´
ıa Jorquera Valero, Manuel Gil P´
erez, and Gregorio Mart´
ınez P´
erez
Faculty of Computer Science, University of Murcia, 30100 Murcia, Spain
Email: {josemaria.jorquera, mgilperez, gregorio}@um.es
Abstract—5G networks intend to cover user demands through
multi-party collaborations in a secure and trustworthy manner.
To this end, marketplaces play a pivotal role as enablers for
network service consumers and infrastructure providers to offer,
negotiate, and purchase 5G resources and services. Nevertheless,
marketplaces often do not ensure trustworthy networking by
analyzing the security and trust of their members and offers.
This paper presents a security and trust framework to enable
the selection of reliable third-party providers based on their
history and reputation. In addition, it also introduces a reward
and punishment mechanism to continuously update trust scores
according to security events. Finally, we showcase a real use case
in which the security and trust framework is being applied.
Index Terms—Security, Trust Management, 5G, Trustworthy
Relationships
Tipo de contribuci´
on: Investigaci´
on en desarrollo
I. INTRODUCTION
The fifth generation of mobile networks (5G) entails a
continuously increasing figure of interconnected end-users as
well as relationships among entities. Thereby, 5G resource,
service, and infrastructure providers are constantly working on
innovative solutions to deal with the huge demand from ten-
ants and users for data capacity, bandwidth, network coverage,
and latency. One of the most promising approaches is decen-
tralized marketplaces. A marketplace is normally described as
a centralized or decentralized repository in which thousand of
offers are advertised by resource and infrastructure providers.
The trading of such resources and services is carried out by
consumers who need to cover dynamic requirements and high
Quality-of-Service (QoS); for instance, high speed, low energy
consumption, and automatically satisfy user demands. In this
context, marketplaces boost the generation of service chains
across operators with security and trustworthiness [1].
Conventionally, trust has been employed as a mechanism
capable of determining the trustworthiness level that a trustor
has in a trustee. In this regard, there are multiple models for
determining a trust level such as PKI-based, role-based, or
reputation-based, among others. The latter is one of the most
considered models, since it not only allows the trust to be cal-
culated based on entity’s behavior in previous interactions but
also allows gathering feedback from reliable recommenders
who also interacted with the entity. Yet, reputation-based trust
models also open the door to key challenges to be addressed
whether they want to be one of the technologies that await to
support next-generation solutions (5G and beyond) [2].
Especially, trust models should guarantee a minimum set
of features in order to be integrated with promising 5G
solutions. First and foremost, trust models need to provide
highly dynamic and context-dependence solutions as 5G
ecosystems tend to support scenarios where main actors
scale and migrate flexibly. In this sense, trust models should
continuously collect information from the principal actors
involved in the relationship through automatic mechanisms
triggered by events, rules, time, etc. By means of these
mechanisms, the trust score can be rapidly adapted in real-
time, and in consequence, adjust the participating entities, if
necessary. Besides, trust models should ensure reliable end-to-
end establishments, therefore any intermediate entities must
be analyzed. Another pivotal aspect to be covered is the
automated management following a zero-touch approach. On
the one hand, trust models should leverage tools that enable
the automatization of network and service management via
high-level policies, and artificial intelligence algorithms. On
the other hand, they should empower a flexible integration
with other 5G essential services and their workflows. Lastly,
trust models should fulfill the zero trust principle [3]. It
attempts not to attribute implicit trust to an entity regardless of
whether the trustor and trustee had previously a relationship or
whether both entities belong to the same domain. Therefore,
trust models would avoid utilizing an outdated trust value
without being updated at a given time as well as assign trust-
by-default to entities under our reliable zone.
Despite the progress of literature, more efforts are needed
to enhance the trust and reputation models in 5G scenarios as
prior trust models did not contemplate all requirements set out
above. Furthermore, the trust concept also includes security
aspects, therefore considering security-related features for
profiling actor behavior would help to detect feasible threats
as well as to have a broader view when determining trust. In
this regard, this paper presents a security and trust framework
capable of ensuring a trustworthy network inside of a decen-
tralized marketplace. In particular, such a framework enables
consumers to establish trustworthy end-to-end connections
across domains, at the same time it covers the aforementioned
challenges. To this end, the security and trust framework
analyzes a set of product offers, published by resource and
infrastructure providers in a decentralized marketplace, in
order to predict trust scores from previous interactions as well
as recommendations from trustworthy third parties. It should
be pointed out that such a framework is being developed in
the 5GZORRO H2020 European project [4], which enables a
secure, flexible, and distributed multi-stakeholder combination
and composition of resources and services in 5G networks.
Additionally, the security and trust framework also introduces
a reward and punishment mechanism to continuously update
trust levels from multiple security events obtained in real-time.
Finally, the paper introduces a real use case to demonstrate
how the framework is integrated with other 5G services and
helps to perform a smart resource and service discovery.
arXiv:2210.11517v1 [cs.CR] 20 Oct 2022