Dual Threats in RIS-Aided RF-UOWC Mixed Networks: Secrecy Performance Analysis under Simultaneous RF and UOWC Eavesdropping

TL;DR

This paper addresses secrecy in a RIS-aided dual-hop RF-UOWC network under three eavesdropping scenarios by modeling RF channels with $\alpha-\mu$ fading and UOWC channels with the $mEGG$ distribution. It develops closed-form expressions for average secrecy capacity, secrecy outage probability, the probability of strictly positive secrecy capacity, and effective secrecy throughput, complemented by high-$SNR$ asymptotics and Monte Carlo validation. The results demonstrate substantial secrecy gains from deploying RIS in both hops and reveal the influence of fading, pointing errors, turbulence, and salinity on security. The framework provides actionable insights for designing secure, RIS-enabled hybrid RF-UOWC systems in 6G contexts, including practical considerations like detection techniques and eavesdropper RIS configurations.

Abstract

In the dynamic realm of 6G technology, emphasizing security is essential, particularly for optimizing high-performance communication. A notable strategy involves the use of reconfigurable intelligent surfaces (RISs), an emerging and cost-efficient technology aimed at fortifying incoming signals, broadening coverage, and ultimately improving the overall performance of systems. In this paper, we introduce a comprehensive framework to analyze the secrecy performance of an RIS-assisted mixed radio frequency (RF) - underwater optical wireless communication (UOWC) network. Here, all the RF links undergo alpha-mu fading distribution, whereas the UOWC links experience a mixture of Exponential Generalized Gamma distribution. Specifically, we examine three potential eavesdropping situations: 1) eavesdropping on the RF link, 2) eavesdropping on the UOWC link, and 3) a simultaneous eavesdropping attack affecting both RF and UOWC links. To achieve this, we derive novel mathematical expressions such as average secrecy capacity, secrecy outage probability, strictly positive secrecy capacity, and effective secrecy throughput in closed form. Using these derived expressions, we carry out an investigation to assess the influences of fading parameters, pointing errors, receiver detection technique, underwater turbulence severity, and water salinity on the system. Furthermore, our study investigates the significance of RIS in improving secrecy performance due to the proposed model. To provide deeper insights, we also perform asymptotic analysis for the high signal-to-noise region. Finally, to verify our analytical results, we conduct Monte Carlo simulation using a computer-based technique.

Figures (17)

• Figure 1: The system model incorporates a source ($\mathbf{S}$), a relay ($\mathcal{R}$), a destination user ($\mathcal{D}$), and two eavesdroppers ($\mathcal{E}1$ and $\mathcal{E}2$), along with four RISs ($I_{\mathcal{R}}$, $I_{\mathcal{D}}$, $I_{\mathcal{E}1}$, and $I_{\mathcal{E}2}$).
• Figure 2: The $SOP^{I}$ versus $\Omega_{t}$ for chosen values of $N_{t}$.
• Figure 3: The $SOP^{I}$ versus $\Omega_{t}$ for chosen values of $N_{d}$ and $\Omega_{e_{s}}$.
• Figure 4: The $SOP^{I}$ versus $\Omega_{t}$ for chosen values of $N_{e_{s}}$ and $\phi_{t,r}$.
• Figure 5: The $SPSC^{III}$ versus $\Omega_{t}$ for chosen values of $N_{e_{r}}$ and $\Omega_{r_{e_{r}}}$.

Theorems & Definitions (10)

• remark 1
• remark 2
• theorem 1
• remark 3
• theorem 2
• remark 4
• theorem 3
• theorem 4
• theorem 5
• theorem 6