The adoption of Implantable Internet of Things Medical Devices (IoTMD) has revolutionized chronic disease management by enabling continuous monitoring and real-time data transmission, allowing patients to optimize treatment strategies. However, these advancements come with significant security risks, as IoTMD systems remain vulnerable to cyber threats that could compromise patient data and device functionality. Addressing this challenge, this study evaluates the fine-tuning performance of various lightweight Large Language Models (LLMs) for anomaly detection in IoTMD-enabled diabetes management control systems (DMCS). Among the evaluated models, LLaMA 3.2 1B-Instruct, fine-tuned with Low-Rank Adaptation (LoRA), achieves the highest performance, with 99.91% accuracy, perfect precision (100.00%), and a false positive rate of 0%. Comparative analysis against other lightweight LLMs-GPT-2, Phi-1 (1.3B), and Gemma 2B-Instruct-as well as traditional deep learning models such as IL-MLP, IL-CNN, FL-MLP, and FL-CNN, highlights the superior adaptability and robustness of transformer-based architectures in anomaly detection. These findings demonstrate the effectiveness of LLMs in securing IoTMD systems, providing a powerful solution for mitigating cyber threats while ensuring system reliability. The results underscore the potential of LLM-based anomaly detection in strengthening IoTMD cybersecurity, paving the way for safer and more reliable implantable medical devices in modern healthcare settings.

The 5G-AKA protocol, a foundational component for 5G network authentication, has been found vulnerable to various security threats, including linkability attacks that compromise user privacy. To address these vulnerabilities, we previously proposed the 5G-AKA-Forward Secrecy (5G-AKA-FS) protocol, which introduces an ephemeral key pair within the home network (HN) to support forward secrecy and prevent linkability attacks. However, a re-evaluation uncovered minor errors in the initial BAN-logic verification and highlighted the need for more rigorous security validation using formal methods. In this paper, we correct the BAN-logic verification and advance the formal security analysis by applying an extended SVO logic, which was adopted as it provides a higher level of verification compared to BAN logic, incorporating a new axiom specifically for forward secrecy. Additionally, we enhance the ProVerif analysis by employing a stronger adversarial model. These refinements in formal verification validate the security and reliability of 5G-AKA-FS, ensuring its resilience against advanced attacks. Our findings offer a comprehensive reference for future security protocol verification in 5G networks

The rapid evolution of mobile and optical communication technologies is driving the transition from 5G to 6G networks. This transition inevitably brings about changes in authentication scenarios, as new security demands emerge that go beyond the capabilities of existing frameworks. Therefore, it is necessary to address these evolving requirements and the associated key challenges: ensuring Perfect Forward Secrecy (PFS) to protect communications even if long-term keys are compromised and integrating Post-Quantum Cryptography (PQC) techniques to defend against the threats posed by quantum computing. These are essential for both radio and optical communications, which are foundational elements of future 6G infrastructures. The DMRN Protocol, introduced in 2022, represents a major advancement by offering both PFS and PQC while maintaining compatibility with existing 3rd Generation Partnership Project (3GPP) standards. Given the looming quantum-era challenges, it is imperative to analyze the protocol’s security architecture through formal verification. Accordingly, we formally analyze the DMRN Protocol using SVO logic and ProVerif to assess its effectiveness in mitigating attack vectors, such as malicious or compromised serving networks (SNs) and home network (HN) masquerading. Our research found that the DMRN Protocol has vulnerabilities in key areas such as mutual authentication and key exchange. In light of these findings, our study provides critical insights into the design of secure and quantum-safe authentication protocols for the transition to 6G networks. Furthermore, by identifying the vulnerabilities in and discussing countermeasures to address the DMRN Protocol, this study lays the groundwork for the future standardization of secure 6G Authentication and Key Agreement protocols.

Roaming in 5G networks enables seamless global mobility but also introduces significant security risks due to legacy protocol dependencies, uneven Security Edge Protection Proxy (SEPP) deployment, and the dynamic nature of inter-Public Land Mobile Network (inter-PLMN) signaling. Traditional rule-based defenses are inadequate for protecting cloud-native 5G core networks, particularly as roaming expands into enterprise and Internet of Things (IoT) domains. This work addresses these challenges by designing a scalable 5G Standalone testbed, generating the first intrusion detection dataset specifically tailored to roaming threats, and proposing a deep learning based intrusion detection framework for cloud-native environments. Six deep learning models including Multilayer Perceptron (MLP), one-dimensional Convolutional Neural Network (1D CNN), Autoencoder (AE), Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) were evaluated on the dataset using both weighted and balanced metrics to account for strong class imbalance. While all models achieved over 99% accuracy, recurrent architectures such as GRU and LSTM outperformed others in balanced accuracy and macro-level evaluation, demonstrating superior effectiveness in detecting rare but high-impact attacks. These results confirm the importance of sequence-aware Artificial Intelligence (AI) models for securing roaming scenarios, where transient and context-dependent threats are common. The proposed framework provides a foundation for intelligent, adaptive intrusion detection in 5G and offers a path toward resilient security in Beyond 5G and 6G networks.

The 5th Generation Mobile Communication (5G) plays a significant role in the Fourth Industrial Revolution (4IR), facilitating significant improvements and innovations in various fields. The 3rd Generation Partnership Project (3GPP) is currently standardizing the Authentication and Key Management for Application (AKMA) system for the 5G convergence applications (5G cAPPs). The Transport Layer Security (TLS) is recommended as the application-specific Ua* protocol between User Equipment (UE) and Application Function (AF) to securely transmit the AKMA identifiers of UE as well as guarantee traffic protection. Among TLS protocols, session resumption in TLS 1.2 and the Pre-Shared Key (PSK) modes of TLS 1.3 are particularly desirable for Ua*. Unfortunately, the integration of PSK options of TLS 1.3, namely PSK-only, PSK-(EC)DHE, and 0-RTT (0 Round-Trip Time) modes, with AKMA has not yet been thoroughly investigated; hence, security, performance, compatibility, and effectiveness remain uncertain. In response, this paper explores the integration of the TLS 1.3 PSK options with AKMA and investigates the said metrics by conducting formal security verification and emulating exemplary applications. According to the formal verification and experimental results, the PSK-(EC)DH mode shows a security strength trade-off with efficiency. On the one hand, the 0-RTT mode demonstrates better efficiency but exhibits drawbacks on forward secrecy and replay attacks. The result suggests that 0-RTT mode has to be approved to ensure seamless integration of the TLS 1.3 PSK option with AKMA. In addition, adjustment on the AKMA architecture is also imperative to enhance security level.