Computation-Efficient Reflection Coefficient Design for Graphene-Based RIS in Wireless Communications

Reconfigurable intelligent surface (RIS) has recently been in the spotlight as a key enabler for the sixth generation (6G) of wireless communication system. The essence of RIS-assisted wireless communications is to decide how to reconfigure RIS (or control reflection coefficients of unit cells) according to the given wireless environments. In particular, since an RIS is usually composed of multiple unit cells, and the amplitude and phase of a signal reflected by each unit cell are coupled, finding jointly optimal reflection coefficients regarding such aspects requires rich computational resources. In the past, most of the preceding studies builds their reflection coefficient models under less realistic assumptions to address such a challenge. To overcome, this study models a practical and tractable reflection coefficient and proposes a computation-efficient method of near-optimal reflection coefficient design of graphene-based RIS unit cells. The method basically leverages a greedy algorithm, thereby obtaining an near-optimal reflection coefficient without multiple computing iterations. To further gain practicality, we propose a quantized reflection coefficient designing method, and show that 1-bit quantization achieves significant performance, which is close to that without quantization. Our numerical results demonstrate the effectiveness of the proposed algorithm by comparing with several benchmark schemes that require rich computational resources.