Physical Layer Security in 6-th Generation Mobile Networks

The sixth generation (6G) of mobile networks envisions a hyper-connected, intelligent, and immersive digital ecosystem in which also human body will become a node of the network. This ambitious vision, as outlined in the IMT-2030 framework, imposes exceptionally stringent requirements on 6G networks, particularly in achieving hyper-low latency and enabling the seamless integration and modeling of the human body as a functional node within the network. To fulfill these demanding requirements, the adoption of novel transmission technologies is essential. Among the most promising candidates for enabling hyper-low latency communication are All-Optical Processing (AOP)-based Optical Communication via Fiber (OCvF) and Terahertz (THz) Communications. In parallel, Wireless Body Area Networks (WBANs) are envisioned to bridge in-body and on-body devices with the broader 6G infrastructure. Within the WBAN domain, emerging paradigms such as Molecular Communications (MCs), Galvanic Coupling (GC) Communications, and Visible Light Communications (VLCs) are being developed to acquire and transmit a diverse array of human-body-related data. These technologies aim to support a holistic representation of the human body, enabling its integration as a dynamic and intelligent node within the network. However, these technologies introduce novel vulnerabilities and operational constraints. Traditional encryption techniques may prove excessively time-consuming and thus unsuitable for hyper-low latency transmission systems. Moreover, sensors and devices employed within the WBAN context are typically resource-constrained and not equipped to support complex and computationally intensive security techniques. Finally, considering the recent advancements of generative AI-algorithm in terms of complexity and usage, an additional challenge emerge in WBANs: the need to determine whether signals, data, and multimedia content truly originate from a human body. Physical Layer Security (PLS) emerges as a promising solution to these challenges. By leveraging the inherent characteristics of communication channels, PLS enables fast and lightweight security mechanisms that operate directly at the physical layer. These techniques are particularly well-suited to the computational and energy constraints of resource-limited devices and the stringent latency requirements of next-generation communication systems. This thesis applies Physical Layer Security (PLS) techniques for confidentiality and authentication to address the security challenges posed by emerging transmission technologies in 6G networks. A unifying theme across all contributions is the need to secure communication systems that are both latency-sensitive and resource-constrained, particularly in the context of human-centric networking enabled by WBANs. The research spans multiple layers of the 6G architecture, from high-speed backbone technologies to body-level communications, and includes: -Confidentiality for AOP-based OCvF and THz Communications These technologies are key enablers of ultra-low latency transmission in 6G, and the proposed PLS techniques ensure data protection without compromising speed. - Confidentiality of VLC-based Implantable Medical Devices (IMD). Within WBANs, VLC offers a promising channel for intra-body communication. The thesis asses reliability and security for VLC-based implantable devices. -Authentication in GC Coupling communications in WBANs. GC is a low-power technique for on-body data exchange. The work introduces PLS-based authentication methods that respect the energy and computational limitations of wearable devices. -Security threats modeling of diffusion-based MCs in WBANs. MCs represent a novel paradigm for intra-body data transmission. This contribution identifies and models potential vulnerabilities, laying the groundwork for future PLS applications. -Authentication of AI-generated content in WBANs}. As generative AI becomes increasingly capable of mimicking human signals and behavior, this work proposes methods to distinguish genuine human-originated data from synthetic content, ensuring trust in human-body-as-a-node systems.