Abstract
MU-MIMO systems today are limited by the tradeoff between capacity and cost. In this paper, we propose an artificial intelligence-based communication system as a new approach with potency to eliminate these tradeoff. Metasurface as a reconfigurable intelligent surface chosen to be the main transmitting component based on its capability to process software-friendly data format, which possibly increases flexibility on various multi-user cases. Based on simulation results, we show that the proposed system is advantageous in terms of significant design simplification with auto-encoder based deep learning systems.
| Original language | English |
|---|---|
| Pages (from-to) | 18-31 |
| Number of pages | 14 |
| Journal | Journal of Korean Institute of Communications and Information Sciences |
| Volume | 46 |
| Issue number | 1 |
| DOIs | |
| State | Published - 1 Jan 2021 |
Keywords
- Artificial intelligence
- Deep learning
- Metasurface
- Reconfigurable intelligent surface