Hexa-X and data protection evolution in 6G

6G security challenges

Mobile networks have a sophisticated, standardized security architecture which has rarely been subject to high impact cyber-attacks as the generations have evolved. However, as mobile networks provide an enabling function across critical infrastructures, the threat landscape is evolving and becoming more complex. As society and industry become increasingly dependent on communication networks, these networks need to develop into truly trustworthy and resilient systems.

Below we present the security challenges identified in early 6G research, including in the EU 6G flagship research project Hexa-X.

Cloudification: The network decomposition paradigm, virtualization, cloudification, and distributed deployment result in an increased number of actors participating directly or indirectly in mobile systems. Therefore, more consideration of the trust model is needed, both for vertical trust in the cloud stack and horizontal trust between cloud platforms and service providers.

Trustworthy AI: Our trust in technology is dependent on our understanding of how it works. This requires understanding why an artificial intelligence (AI) model makes decisions the way it does. Confidence in an AI agent doing what it is intended to do is critical when we consider the increased coupling of AI technology with 6G components. Trustworthy AI strives to achieve explainability as well as fairness, robustness, privacy, and transparency. Therefore, we should ensure the trustworthiness of AI components used in 6G systems and architecture.

End user diversity: Novel applications in 6G will result in an increase in both the number and diversity of end user devices. 6G as a secure network platform should address an expanded set of security requirements introduced by different devices to ensure that all devices work seamlessly in a secure manner.

Privacy: In the data-driven world of 6G architecture, AI models will have access to more data, including potentially sensitive or personal information. Thus, there is a need to implement data protection features in the 6G architecture from the very beginning, which can be activated when new, sensitive, or personal data appears.

End-to-end security: Sound security mechanisms already exist for protecting data in transit, whereas more emphasis needs to be put on protecting data being processed or data being stored to ensure end-to-end security in 6G. Oblivious techniques, confidential computing, homomorphic encryption, and privacy-based identifiers can be applied to both 6G network services and components.

Hexa-X

The Hexa-X European 6G research project has just been completed. The results from the project are expected to provide a solid basis for a future global, standardized mobile network. Hexa-X envisions a future 6G network, which enables seamless interactions between the human (senses, body network intelligence, interactions), physical (devices, objects, processes), and digital (connections, information, communications) worlds. New societal and technological demands lead 6G research to explore new use cases, new architectural components, and new deployment solutions.

Hexa-X based its targets using both key performance indicators and newly introduced key value indicators :

• digital inclusion
• trustworthiness
• sustainability

Security, privacy, and trust have been high on the agenda since the development of 2G. Data usage has increased across generations, from 3G to 5G, putting more and more focus on protecting data. The fundamental security mechanisms in current networks – like secure identities, access control mechanisms, and secure interfaces between architectural components – are still relevant in 6G. To deliver the required level of security, privacy, and trust as networks and their use evolve, Hexa-X has identified a set of 6G security technology enablers.

Overview of 6G security enablers proposed by Hexa-X

Hexa-X Security Enablers Source: Hexa-X D1.3

Trust foundations: Secure digital identities play an essential role in building trust. Secure identities are required for communication in both the connectivity layer and application layer among several distributed entities in mobile networks. Confidential computing (CC) is a secure computation model, which assures users that their environment is secure when they run applications on virtualized environments benefiting from hardware security offerings. It provides hardware enforced isolation with memory encryption through enclave technologies and uses advanced trusted execution environments (TEEs) to build enclaves. The Root-of-Trust (RoT) mechanism forms the foundation of confidential computing, and it leverages the path to secure identities and protocols which depends on establishing trusted identities for infrastructure, connectivity, devices, edge, and network slicing functions.

AI/ML assurance and defense: We expect to see pervasive AI/ML functionalities in 6G in several areas. AI/ML can be used as a tool to enhance functionalities and services, including improving, automating, and optimizing security. AI/ML emerges as a key ally for boosting security assurance and defense enabler solutions against cyberattacks. Intelligent security monitoring and management is required especially to address flexible cloud deployment patterns. In addition, collaborative learning approaches pave the way for conforming data protection legislation as they provide joint model creation without transferring data from distributed data sources. While AI/ML is a powerful tool, the scrutinization of its security and privacy aspects is vital to provide holistic trustworthiness for 6G services. AI/ML systems should be robust to adversarial interventions so that any malicious attempts do not affect the outputs of AI/ML systems. . Ericsson plays a key role , providing technical contributions, leading AI/ML security and privacy discussions, and acting as a liaison between work packages.

Privacy enhancing technologies: Privacy enhancing technologies refer to a set of building blocks that can be used to achieve privacy in communications and computations, for example, differential privacy, homomorphic encryption, and secure multi-party computations. Since each generation of mobile network technologies aims to improve privacy, and with the increasing AI/ML integration in 6G use cases, some of these new technologies might be embraced to meet the privacy requirements that arise in 6G. .

• Differential privacy (DP) is a data anonymization technique to protect the privacy of individuals while analyzing data. It works by adding a small amount of random noise to the data, which helps to make it difficult for anyone to identify an individual's personal information. It is widely used to protect the privacy of the individuals whose information is in a dataset. In the ML context, DP is rigorously addressing membership inference attacks (given a data record and black-box access to a model, if the record was in the model's training dataset).
• Homomorphic encryption (HE) refers to a form of encryption that enables the computation of ciphertext without accessing cleartext data. The resulting computation persists in encrypted form until the keyholder decrypts the result. HE accomplishes this with operations such as addition and multiplication that can be used as the basis for more complex arbitrary functions.
• Secure multiparty computation (SMC) is another cryptographic method allowing parties to jointly compute a function of their sensitive inputs without disclosing their inputs. Some of the approaches and primitives used to realize SMC are Yao’s Garbled circuits, Goldreich, Micali and Wigderson (GMW), oblivious transfer, secret sharing, and homomorphic encryption.

Distributed ledger technologies: The key characteristics of distributed ledger technologies (DLTs), which pave the way for services among distributed actors, are identified as transparency, immutability, non-repudiation, proof of provenance, integrity, and pseudonymity. DLT has the potential to provide distributed trust different stakeholders in a multitenant/multi-domain environment. Hexa-X explored the possible DLT use cases for 6G systems.  

Quantum security: Quantum computing's security implications are emerging as a key 6G technology focus". Quantum-resistant Cryptography, aka Post-quantum Cryptography, is an important research topic to consider in the 6G security landscape aiming to provide cryptographic algorithms which are secure against both quantum and classical computers. These algorithms and their implications in network protocols and related security procedures, like key distribution, should be considered in the design of next-generation networks.

Physical layer security: While cryptography will still be a fundamental aspect of security in 6G, the research community is investigating alternative, and possibly more efficient methodologies to avoid some complexities of traditional methods, for example, to perform the encryption and decryption operations on each data packet or power consumption which is detrimental for IoT. To tackle these complexities, Physical layer security (PLS) approaches can complement existing mechanisms. PLS has emerged principally to exploit the inherent random nature of wireless channels to achieve secure data transmission without a need for cryptographic keys. Hence, security is achieved by exploiting the unique physical properties of the channel or device.

Hexa-X-II: Bringing the building blocks together

The new 6G research flagship project, Hexa-X-II, was kicked off in January 2023 with the main objective of blueprinting a sustainable, inclusive, and trustworthy 6G platform using Hexa-X's identified building blocks.As expected, some of the key research and innovation areas are . These aspects will be addressed during the design of the 6G platform blueprint end-to-end system validation. Security and privacy related objectives include the risk assessment covering security and privacy threat analysis as well as threat impact, and mitigation technologies to address the threats. The project is in its early stages, and the first public deliverables do not address the security topics. Upcoming deliverables will include some of the security and privacy technologies such as distributed and trustworthy AI, quantum-safe crypto, distributed ledgers, remote attestation, and context awareness. Seeing these technologies in the list of topics for 6G security is no surprise because they have already been identified in Hexa-X as mentioned. Since AI is a very promising tool that can be used for other purposes in addition to enhancing the security and privacy of the system, the study of AI trustworthiness has also been prioritized in Hexa-X-II with a focus on sustainable and trustworthy AI/ML based control systems. , concrete views and architectural options will become more visible.

Advancements in security technologies will help us address 6G security and privacy challenges that cannot be handled solely by classical techniques, enabling us to design secure, intelligent, and reliable 6G systems.

Learn more about Hexa-X, 6G use cases, and privacy enhancing technologies

Hexa-X deliverables can be found at Deliverables - Hexa-X, visit the project website 

Read our previous blog posts on Hexa-X: Hexa-X – The joint European initiative to shape 6G and Hexa-X: 6G technology and its evolution so far

Discover how 6G can contribute to an efficient, human-friendly, sustainable society through ever-present intelligent communication: Taking 5G to 6G