Abstracts Communications and the Arts

A Dynamic Subarray Structure in Reconfigurable Intelligent Surfaces for TeraHertz Communication Systems

by Yicong Liu

Reconfigurable Intelligent Surfaces (RIS) have recently gained popularity as a TeraHertz (THz) multi-input multi-output (MIMO) communication solution. THz range (0.1-10 THz) is critical for ubiquitous wireless communications. THz frequencies, in particular, promise to provide sufficient spectrum, data rates exceeding a hundred gigabits per second (Gbps), huge interconnectedness, denser networks, and highly secure communications. While a higher carrier frequency entails a larger communication path loss. To lower it, the primary option is to improve the reflecting signal, and RIS has several benefits that make it a good candidate for THz communication. When the size of a Base Station (BS) antenna is constant, a larger RIS array can boost spectral efficiency, according to earlier studies. However, if the RIS is significantly larger than the BS, the BS beam will not cover the entire RIS array. So there is a trade-off between the spectral efficiency and the number of RIS elements where RIS utilization is maximal. Different from finding out the appropriate amount of RIS elements in a system, optimizing the RIS reflecting pattern to improve the coverage of BS beams and raise the system spectral efficiency is another technical route. Based on that, on the one hand, we separate the whole RIS array into some of subarrays, and each subarray aligns to only one corresponding user, making RIS-subarray-user pairs in a multiuser THz communication system. On the other hand, a system with a general RIS structure where the RIS reflects beams to all users without subarray is considered as a comparison. Each RIS subarray independently reflects the beam to the corresponding user in this system. We settle it because if the size of RIS is much larger than the BS array, the beams from BS to RIS will hardly cover all RIS elements, and the utilization rate of RIS will be lower down.