Coverage Analysis for mmWave Networks With Reflective and Transmissive Intelligent Surfaces
Reconfigurable intelligent surfaces (RISs) have been proposed to enhance coverage performance in millimeter-wave bands by providing alternative links between access points and user devices in non-line-of-sight (NLOS) scenarios. However, the previously-studied reflective RISs (R-RISs) only offer 180° coverage, with most studies focusing on links with one R-RIS. Recently, transmissive-reflective RISs (T-RISs) that can provide 360° coverage have been proposed. In order to understand performance limits of both types of RISs, stochastic geometry is employed to analyze connection probability when R-RISs and T-RISs are used with single-RIS and multi-RIS links. The connection probability for single-RIS links and an upper bound on connection probability for multi-RIS links are derived with sparse obstacle distributions where independence of line-of-sight (LOS) statuses of different links can be assumed. The theoretical analysis is validated by simulations. Additionally, a comparison is provided between single-RIS and two-RIS links, as well as between R-RISs and T-RISs. Numerical evaluation using Nakagami fading propagation and a sectored antenna model shows that single-RIS links offer substantial coverage improvement for shorter-distance communications, whereas two-RIS links are more effective for longer-range communications. Moreover, numerical results demonstrate that, under the same model, T-RISs exhibit significantly improved coverage compared to R-RISs, especially with denser obstacle distribution.
Authors
- Douglas M. Blough (ORCID: https://orcid.org/0000-0002-0803-7647)
- Jingyuan Zhang (ORCID: https://orcid.org/0009-0008-2150-9902)
Institutions
- Georgia Institute of Technology (US)
Publication Details
- Journal
- IEEE Transactions on Wireless Communications
- Published
- 2024-10-31
- DOI
- https://doi.org/10.1109/twc.2024.3486091
- Citations
- 2
- Primary Topic
- Advanced Wireless Communication Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.42