[Paper Review] A Simple Multiple-Access Design for Reconfigurable Intelligent Surface-Aided Systems

Wei Jiang, Hans D. Schotten|arXiv (Cornell University)|Sep 12, 2023

Advanced Wireless Communication Technologies4 citations

TL;DR

This paper proposes Opportunistic Reflection-based Non-Orthogonal Multiple Access (OR-NOMA), a low-complexity, high-performance multiple-access scheme for RIS-aided multi-user systems. By selecting the user with the strongest channel gain for RIS reflection optimization and enabling non-orthogonal transmission with a second user, OR-NOMA achieves near-joint optimization performance with orders-of-magnitude lower complexity, while a random-phase variant avoids CSI overhead and outperforms opportunistic beamforming.

ABSTRACT

This paper focuses on the design of transmission methods and reflection optimization for a wireless system assisted by a single or multiple reconfigurable intelligent surfaces (RISs). The existing techniques are either too complex to implement in practical systems or too inefficient to achieve high performance. To overcome the shortcomings of the existing schemes, we propose a simple but efficient approach based on extit{opportunistic reflection} and extit{non-orthogonal transmission}. The key idea is opportunistically selecting the best user that can reap the maximal gain from the optimally reflected signals via RIS. That is to say, only the channel state information of the best user is used for RIS reflection optimization, which can in turn lower complexity substantially. In addition, the second user is selected to superpose its signal on that of the primary user, where the benefits of non-orthogonal transmission, i.e., high system capacity and improved user fairness, are obtained. Additionally, a simplified variant exploiting random phase shifts is proposed to avoid the high overhead of RIS channel estimation.

Motivation & Objective

• To address the high complexity and inefficiency of existing RIS-aided multiple access schemes in multi-user, multi-RIS systems.
• To reduce the computational burden of RIS reflection optimization by leveraging only the CSI of the best user.
• To enable efficient non-orthogonal multiple access with improved spectral efficiency and user fairness.
• To propose a practical variant using random phase shifts to eliminate the need for costly RIS channel estimation.

Proposed method

• Opportunistic user selection: the user with the strongest effective channel gain is selected for RIS reflection optimization.
• RIS reflection coefficients are optimized based solely on the CSI of the selected user, drastically reducing feedback and computation overhead.
• Non-orthogonal multiple access (NOMA) is employed: the second user superposes its signal on the primary user’s signal for multiplexing gain.
• A simplified variant, OR-NOMA-RP, uses random phase shifts at the RIS to avoid CSI estimation entirely.
• Successive interference cancellation (SIC) is applied at the receiver to decode the two superimposed signals.
• The system is evaluated via Monte Carlo simulations under realistic 3GPP UMi and Nakagami-m fading conditions.

Experimental results

Research questions

• RQ1 Can a low-complexity multiple access scheme achieve near-optimal performance in RIS-aided multi-user systems by exploiting only partial CSI?
• RQ2 How does opportunistic user selection combined with NOMA improve spectral efficiency and fairness compared to orthogonal schemes?
• RQ3 What is the performance-complexity trade-off of OR-NOMA versus joint reflection optimization (JRO) and opportunistic beamforming (OppBF)?
• RQ4 Can random phase shifts at the RIS provide a practical alternative to CSI-based optimization with minimal performance loss?
• RQ5 How does OR-NOMA perform under hardware impairments such as phase noise and channel aging?

Key findings

• OR-NOMA achieves performance within close proximity to joint reflection optimization (JRO), the theoretical optimum, while reducing computational complexity by four orders of magnitude.
• The average CPU runtime for OR-NOMA is 1.43 ms per channel realization, compared to 23,900 ms for JRO, making JRO infeasible for real-time applications.
• In the absence of CSI, OR-NOMA-RP (with random phase shifts) achieves approximately 5 dB SNR gain over OppBF, demonstrating robustness to channel estimation overhead.
• All multi-user schemes outperform single-user transmission due to multiplexing gain, with OR-NOMA significantly surpassing TDMA and OUS.
• The proposed scheme maintains high spectral efficiency and fairness, especially under limited feedback and hardware constraints.
• The performance of OR-NOMA is robust under realistic fading and path-loss models, including 3GPP Urban Micro (UMi) and Nakagami-m fading.