Physical Layer Security in Cooperative NOMA Hybrid VLC/RF Systems

TL;DR

This paper explores physical layer security in a hybrid VLC/RF system using cooperative NOMA, proposing solutions to enhance secrecy rates against eavesdroppers through beamforming, power allocation, and jamming techniques.

Contribution

It introduces novel methods for maximizing secrecy rate in hybrid VLC/RF systems with cooperative NOMA, considering both active and passive eavesdroppers.

Findings

Proposed approaches improve secrecy performance significantly.

Beamforming and jamming enhance security against eavesdroppers.

Numerical results validate the effectiveness of the proposed methods.

Abstract

Integrating visible light communication (VLC) and radio-frequency (RF) networks can improve the performance of communication systems in terms of coverage and data rates. However, adding RF links to VLC networks weakens the secrecy performance due to the broadcast and ubiquitous nature of RF links. This paper studies the physical layer security (PLS) in cooperative non-orthogonal multiple access (CoNOMA) hybrid VLC/RF systems. Consider a VLC system, where two entrusted users close to a VLC access point (AP) help an out-of-coverage legitimate user using RF signals in the presence of an eavesdropper. The AP transmits data to both entrusted users and the legitimate user using the principle of NOMA, where the entrusted users harvest energy from the received light intensity, decode the legitimate user's message, forward it using a RF link, and then decode their messages. It is required to maximize the secrecy rate at the legitimate user under quality-of-service (QoS) constraints using beamforming and DC-bias and power allocation. Different solutions are proposed for both active and passive eavesdropper cases, using semidefinite relaxation, zero-forcing, beamforming, and jamming. Numerical results compare between the different proposed approaches and show how the proposed approaches contribute in improving the secrecy performance of the proposed model.