Wireless Physical Layer Security Insights for Non-orthogonal

Understanding Non-orthogonal Multiple Access (NOMA)

Non-orthogonal multiple access (NOMA) has gained significant attention as a key enabling technology for 5G and beyond wireless networks. Unlike traditional orthogonal multiple access (OMA) schemes, NOMA allows multiple users to share the same time-frequency resources by exploiting the power domain. This unique property of NOMA leads to enhanced spectral efficiency, improved coverage, lower latency, and support for massive connectivity.

There are two main types of NOMA:

1. Power-domain (PD) NOMA: In PD-NOMA, the base station (BS) superimposes the signals of multiple users with different power allocation based on their channel conditions. The near user (NU) with a better channel is assigned lower transmit power, while the far user (FU) with a poorer channel is allocated higher transmit power. The NU first decodes the FU’s signal, performs successive interference cancellation (SIC) to remove it, and then decodes its own signal. The FU directly decodes its signal, treating the NU’s signal as interference.

2. Code-domain (CD) NOMA: CD-NOMA is similar to code-division multiple access (CDMA), where users share the same time-frequency resources but are distinguished by unique non-orthogonal spreading codes with low cross-correlation.

While both uplink and downlink NOMA have been studied, the focus of this article is on downlink PD-NOMA, as it has been more extensively explored by standardization bodies such as 3GPP and IEEE.

Security Challenges in NOMA

Compared to OMA, NOMA faces some critical security risks due to its broadcast nature. Since the BS transmits a superimposed signal containing the data of multiple users, an eavesdropper (Eve) can potentially intercept the information of all users if the NOMA transmission is successfully compromised. Moreover, in NOMA, there is a need to secure the confidential messages of each user from one another, especially in scenarios where the users are untrusted.

To address these security concerns, physical layer security (PLS) techniques have emerged as a promising solution that can complement or even replace traditional cryptography-based approaches. PLS exploits the dynamic features of wireless channels, such as random fading, interference, and noise, to prevent eavesdroppers from decoding the data while ensuring that the legitimate users can decode it successfully.

Security Design Objectives for NOMA

The security design objectives for NOMA can be divided into three main categories:

1. Security against external eavesdroppers: In this scenario, the NU and FU are considered trusted, and the design goal is to secure their messages from an external eavesdropper. The location of the eavesdropper can affect the security performance, as users are protected unequally due to the different power allocation in NOMA.

2. Security against internal eavesdroppers: In this case, there are no external eavesdroppers, but the users are untrusted. The design goal is to secure the information of each user from the others, while ensuring that the SIC operation works normally.

3. Security against both external and internal eavesdroppers: This is the most challenging scenario, where the design must address the security of the NU and FU signals against both external and internal eavesdroppers.

PLS Solutions for Securing NOMA

To address the security challenges in NOMA, various PLS techniques have been proposed in the literature. These include:

1. Beamforming: Beamforming can be used to enhance the signal power at the legitimate users while suppressing it in other directions. However, the beamforming design needs to be carefully modified to ensure that it does not affect the normal SIC operation in NOMA.

2. Artificial Noise (AN) with Beamforming: AN-based techniques, combined with beamforming, can effectively degrade the eavesdropper’s performance, especially when the eavesdropper is closer to the BS than the legitimate users. The performance of these techniques depends on the availability of the eavesdropper’s channel state information (CSI).

3. Power Allocation: Optimizing the power allocation based on the legitimate users’ channel conditions can make the interception of their signals more difficult for the eavesdropper under certain settings.

4. Cooperative Beamforming and Jamming: Cooperative communication, with distributed beamforming and jamming, can enhance the security of NOMA systems by directing the signal towards the desired users while degrading the eavesdropper’s performance.

For the case of internal eavesdropping, where the users are untrusted, the PLS solutions focus on transforming the signals of the NU and FU into another domain, such that the SIC operation can be performed normally, but the eavesdropping users cannot decode the information of their partners.

Merits of PLS in NOMA

Compared to OMA, NOMA offers several advantages when it comes to PLS:

1. Collaborative Multi-user Processing: In NOMA, the signals are not sent separately as in OMA. This allows for collaborative processing of multi-user interference and PLS, enabling the design of user selection, clustering, power allocation, and other parameters based on the quality of service requirements of the legitimate users.

2. Efficient Security in Massive MIMO: In massive MIMO systems, AN-based security techniques face complexity issues. NOMA can help provide secure communication without the need for AN, by exploiting the inter-user interference intelligently.

3. Securing Unicast and Multicast Transmissions: NOMA can be used to secure a unicast message from interception by untrusted multicast receivers, while improving the spectral efficiency compared to OMA.

4. Robustness to Poor Scattering Environments: Most PLS techniques assume a rich scattering environment, which may not always be the case. NOMA with large-scale fading-based security algorithms can provide secure communication even in poor scattering environments.

Challenges and Future Recommendations

While there have been considerable contributions in securing NOMA using PLS, several challenges and open research directions remain:

1. Passive and Active Eavesdropping: Most existing techniques assume a passive eavesdropper, but future networks may face active attacks, such as pilot spoofing, that can disrupt the normal operation of the NOMA system. Designing PLS techniques robust to active attacks is an important research direction.

2. SIC Error Propagation and Imperfect CSI: Security algorithms in NOMA often rely on the assumption of perfect SIC and channel estimation. The effects of imperfect SIC and channel estimation should be considered when designing security algorithms.

3. Artificial Noise Challenges: AN-based techniques can lead to increased peak-to-average power ratio, power wastage, and sensitivity to imperfect CSI. Designing AN-based schemes that address these challenges is necessary.

4. Multi-cell and Integrated Technologies: Securing NOMA in multi-cell scenarios and when integrated with other technologies, such as millimeter-wave, full-duplex, and cognitive radio, requires further investigation.

5. Cross-layer and Hybrid Security Approaches: Exploring cross-layer security designs that optimize physical layer parameters based on upper-layer features, as well as hybrid techniques combining PLS and cryptography, can further enhance the security of NOMA-based systems.

6. IRS-assisted PLS for NOMA: Reconfigurable intelligent surfaces (RIS) can be exploited to enhance PLS against both external and internal eavesdroppers in NOMA networks, which is an emerging research direction.

By addressing these challenges and exploring the recommended research directions, practical and secure NOMA systems can be developed to meet the stringent efficiency and security requirements of 5G and beyond wireless networks.

Conclusion

NOMA and PLS are two promising technologies that, when combined, can enable spectrally efficient, adaptive, and secure wireless communication systems. This article has discussed the key security design requirements for NOMA, the strengths of PLS as a solution, and the merits of employing PLS in NOMA. However, the challenges and future recommendations highlighted in this work need to be investigated further to address the open issues and develop practical secure NOMA systems.

As an experienced IT professional, I believe the insights provided in this article can help guide the development of robust and secure NOMA-based solutions, which will be crucial for the successful deployment of 5G and beyond wireless networks.