[Paper Review] Analysis and Optimization for RIS-Aided Multi-Pair Communications Relying on Statistical CSI

Zhangjie Peng, Tianshu Li+4 more|arXiv (Cornell University)|Jul 22, 2020

Advanced Wireless Communication Technologies17 references4 citations

TL;DR

This paper proposes a genetic algorithm (GA)-based optimization framework for reconfigurable intelligent surface (RIS)-aided multi-pair communications using statistical channel state information (CSI). It derives an approximate achievable rate expression and demonstrates that three quantization bits per RIS element achieve nearly optimal performance compared to continuous phase shifts, with the GA method closely matching the globally optimal solution in sum rate performance.

ABSTRACT

In this paper, we investigate a reconfigurable intelligent surface (RIS) aided multi-pair communication system, in which multi-pair users exchange information via an RIS. We derive an approximate expression of the achievable rate by assuming that statistical channel state information (CSI) is available. A genetic algorithm (GA) to solve the rate maximization problem is proposed as well. In particular, we consider implementations of RISs with continuous phase shifts (CPSs) and discrete phase shifts (DPSs). Simulation results verify the correctness of the obtained results and show that the proposed GA method has almost the same performance as the globally optimal solution. In addition, numerical results show that three quantization bits can achieve a large portion of the sum achievable rate for the CPSs setup.

Motivation & Objective

• To address the challenge of low-rate, high-complexity multi-pair communications in environments with blocked line-of-sight links.
• To develop a practical phase shift optimization method for RIS-aided systems under statistical CSI, which is more feasible than instantaneous CSI.
• To evaluate the performance trade-off between continuous and discrete phase shifts (CPSs vs. DPSs) in terms of achievable sum rate.
• To provide engineering insights on the minimum number of quantization bits required for DPSs to achieve near-optimal performance.

Proposed method

• Derives an approximate closed-form expression for the achievable sum rate under Rician fading channels using statistical CSI.
• Proposes a genetic algorithm (GA) to optimize the phase shifts of RIS elements for both continuous (CPSs) and discrete (DPSs) phase shift configurations.
• Models the RIS-aided multi-pair system with K user pairs, where each pair communicates via an RIS with L reflective elements, using Rician fading for both links.
• Implements the GA with population size 100, 50 generations, crossover and mutation rates set to 0.5 and 0.1, respectively, with convergence threshold $10^{-6}$.
• Uses Monte Carlo simulations to validate the analytical rate expression and compares the GA performance against random phase shifts and exhaustive search.
• Evaluates the impact of quantization bits (B) on sum rate performance, particularly focusing on B=3.

Experimental results

Research questions

• RQ1 What is the achievable rate performance of an RIS-aided multi-pair communication system when only statistical CSI is available?
• RQ2 How does the performance of a genetic algorithm-based phase shift optimization compare to the globally optimal solution in terms of sum rate?
• RQ3 What is the minimum number of quantization bits per RIS element required to achieve a large portion of the rate performance of continuous phase shifts?
• RQ4 How does the Rician factor affect the sum rate performance and the gap between optimized and random phase shift schemes?

Key findings

• The derived analytical expression for the achievable sum rate shows strong agreement with Monte Carlo simulation results, validating the theoretical framework.
• The proposed GA-based optimization achieves sum rate performance nearly identical to the globally optimal solution obtained via exhaustive search, especially under low SNR and moderate Rician factors.
• With only three quantization bits per RIS element, the system achieves approximately 95% of the sum rate performance of continuous phase shifts, indicating high spectral efficiency with low hardware complexity.
• The sum achievable rate increases with the number of RIS elements L and signal-to-noise ratio (SNR), as expected, with a notable performance gain observed at higher SNR.
• As the Rician factor increases, the performance gap between the GA-optimized and random phase shift schemes converges to a fixed value, indicating that LOS dominance reduces the benefit of optimization.