José Dias

Network slicing is a core enabler of multi-tenant 5th Generation (5G) architectures, allowing heterogeneous services to coexist over shared infrastructure. However, ensuring effective isolation between slices remains a critical security challenge, as failures may enable cross-slice interference, data leakage, or cascading service disruption. This article analyses security vulnerabilities affecting 5G network slicing from a risk-oriented perspective, with particular emphasis on isolation weaknesses across orchestration, virtualization, network, and interface layers. Due to the technical immaturity and instability of current open-source slicing platforms, experimental validation of security mechanisms proved infeasible. These limitations are therefore treated as empirical evidence informing a structured vulnerability taxonomy and a qualitative risk assessment grounded in confidentiality, integrity, and availability. Building on this analysis, the article proposes a conceptual security framework that integrates defence-in-depth, zero-trust principles, continuous monitoring, and adaptive response mechanisms to enforce isolation dynamically. Aligned with established standards and regulatory references, the framework provides a coherent theoretical foundation for future experimental validation and the secure design of resilient 5G network slicing architectures.

This paper explores the security challenges associated with network slicing in 5th Generation (5G) networks, a technology that enables the creation of virtual networks tailored to different use cases. This study contributes to network slicing research efforts by providing a comprehensive classification of attacks aligned with the architectural layers of 5G, complemented by practical mitigation approaches suitable for multi-tenant environments. The classification depicts specific attacks and categorizes vulnerabilities across layers such as orchestration, virtualization, and inter-slice communication. Additionally, mitigation strategies are discussed, emphasizing the importance of real-time monitoring and robust access controls. The proposed classification aims to support the development of advanced security mechanisms, including risk assessment models and automated mitigation strategies, tailored to the dynamic and heterogeneous nature of 5G slicing. The findings highlight the need for layered defenses, AI-driven monitoring, and architectural isolation as critical components to enhance the resilience of 5G slicing deployments.

The General Data Protection Regulation (GDPR) is the regulation that determines the directives inherent to the collection, processing, and protection of personal data in European Union (EU) countries. It was implemented in May 2018 and over the past few years, several public and private companies have been affected by serious penalties. With more than 1500 fines already registered, it is important to have an analysis and insights about them. This paper proposes a detailed analysis of the public records of fines under GDPR, understanding the average fines imposed, the main causes for their application and how they have evolved over time. It is also intended to understand the most affected sectors and point ways to mitigate these penalties. It is concluded that fines under GDPR have an increasing trend over time, both in number of fines and in value, with Industry and Commerce & Media, Telecoms and Broadcasting being the most affected sectors. DOI: https://doi.org/10.54663/2182-9306.2023.sn12.42-58