Reconfigurable Intelligent Surface-Aided OFDM Wireless Communications: Hardware Aspects of Reflection Optimization Methods

Abstract

A Reconfigurable Intelligent Surface (RIS) is a passive system node, envisioned as a new physical layer technology in sixth-generation (6G) infrastructure. A RIS enables smart propagation environments by tuning the signal reflection in real time. Reflection optimization is an active field of research, regarding the RIS integration into wireless networks. A real-time response is required, while the unit-modulus constraint on the phase shift introduced by RIS elements makes the optimization problem even more challenging. In this paper, both low-complexity and efficient, in terms of achievable rate, passive beamforming methods are evaluated over a frequency-selective fading channel. Practical aspects of a RIS, including binary phase shifts with unbalanced amplitudes and mutual coupling, are considered. A hardware-oriented reflection optimization method is proposed and is implemented on a Zynq UltraScale+ multiprocessor system-on-a-chip (MPSoC) device. The proposed architecture leads to an extremely low execution time, useful especially for high mobility scenarios, in which coherence time is a tough constraint.