Introduction
As wireless networks transition toward 5G and beyond, Non-Orthogonal Multiple Access (NOMA) has become a cornerstone technology. By allowing multiple users to share the same time and frequency resources through power-domain multiplexing, NOMA significantly enhances spectral efficiency. However, this broadcasting nature makes NOMA inherently vulnerable to eavesdropping. While traditional security relies on upper-layer encryption, Physical Layer Security (PLS) offers a robust alternative by exploiting the inherent characteristics of the wireless channel to ensure secure communication.
System Model
The research focuses on a downlink NOMA scenario involving a Base Station (BS), a "strong" user with favorable channel conditions, and a "weak" user with poorer conditions. A malicious eavesdropper attempts to intercept the signals intended for these legitimate users. To counter this threat, a cluster of $K$ trusted, half-duplex relays is introduced to assist the BS.
The goal is to maximize the secrecy rate region—the set of achievable secure data rates for both legitimate users—by optimizing the relaying strategy and beamforming design.
Relaying Strategies
The study evaluates three distinct relaying schemes, each employing specialized beamforming to mitigate the eavesdropper's impact:
- Cooperative Jamming (CJ): Relays transmit a noise signal specifically designed to fall into the null space of the legitimate users' channels. This effectively degrades the eavesdropper's reception without interfering with the intended users.
- Decode-and-Forward (DF): In a two-phase process, relays first decode the BS's broadcast and then re-transmit the messages using beamforming that maximizes the signal at legitimate users while nulling it at the eavesdropper.
- Amplify-and-Forward (AF): Similar to DF, but relays simply amplify and re-transmit the received signal (including noise) using optimized beamforming vectors.
The core innovation lies in the derivation of optimal beamforming solutions for these schemes, often utilizing generalized eigenvector analysis to coordinate the $K$ relays effectively.
Impact of Network Geometry
A critical insight from the research is that no single relaying scheme is universally superior. The performance of each strategy is highly sensitive to the network's spatial configuration, particularly the distance of the relays and the eavesdropper from the BS.
As shown in the simulations, the Cooperative Jamming (CJ) scheme often outperforms the more complex DF and AF methods when relays are positioned far from the BS. Conversely, DF and AF tend to be more effective when relays are closer to the source. Furthermore, in scenarios where the eavesdropper is closer to the BS than the legitimate users—rendering direct transmission's secrecy rate zero—the presence of relays becomes the sole enabler of secure communication.
Key Results and Conclusion
Numerical simulations demonstrate significant secrecy gains when employing trusted relays compared to direct transmission. Key findings include:
- Relay Scaling: The performance of the AF scheme is notably more sensitive to the number of relays ($K$) than CJ or DF.
- Robustness of CJ: The CJ scheme shows a monotonic increase in performance as relays move further from the BS toward the users and eavesdropper.
- Null-Space Advantage: With at least three relays ($K \ge 3$), the system can consistently find beamforming vectors that cancel the signal at the eavesdropper's location.
While the study assumes perfect Channel State Information (CSI) and half-duplex operation, it provides a foundational framework for securing NOMA systems. Future work may address more realistic constraints, such as imperfect CSI, full-duplex relays, and scaling to massive NOMA environments.
Summary updated