A Deep Learning Based Modeling of Reconfigurable Intelligent Surface Assisted Wireless Communications for Phase Shift Configuration

Baoling Sheen,Jin Yang,Md Moin Uddin Chowdhury,Xianglong Feng

doi:10.1109/ojcoms.2021.3050119

Abstract

Reconfigurable Intelligent Surface (RIS) has emerged as a promising technology in wireless networks to achieve high spectrum and energy efficiency. RIS typically comprises a large number of low-cost nearly passive elements that can smartly interact with the impinging electromagnetic waves for performance enhancement. However, optimally configuring massive number of RIS elements remains a challenge. In this article, we present a machine learning (ML) based modeling approach that learns the interactions between the phase shifts of the RIS elements and receiver (Rx) location attributes and uses them to predict the achievable rate directly without using channel state information (CSI) or received pilots. Once learned, our model can be used to predict optimal RIS configuration for any new receiver locations in the same wireless network. We leverage deep learning (DL) techniques to build our model and study its performance and robustness. Simulation results demonstrate that the proposed DL model can recommend near-optimal RIS configurations for test receiver locations which achieved close to an upper bound performance that assumes perfect channel knowledge. Our DL model was trained using less than 2% of the total receiver locations. This promising result represents great potential in developing a practical solution for the optimal phase shifts of RIS elements without requesting CSI from the wireless network infrastructure.

Highlights

R IS (Reconfigurable Intelligent Surfaces) has been envisioned as a promising technology to reduce the energy consumption and improve the communication performance by artificially reconfiguring the propagation environment of electromagnetic (EM) waves
We propose a new deep learning (DL)-based approach which aims to learn the characteristics of the channel variations in the propagation environment between the communication devices
Note that our proposed approach is machine learning (ML)-based, we describe the channel model and problem formulation based on traditional communication theory-based mathematic formulas first in Sections II and III; later in this article, we describe how the relationships can be learned without using channel state information

Summary

Introduction

R IS (Reconfigurable Intelligent Surfaces) has been envisioned as a promising technology to reduce the energy consumption and improve the communication performance by artificially reconfiguring the propagation environment of electromagnetic (EM) waves. Some fundamental characteristics that make RIS different from current available technologies as pointed out in [2], [3] include the unique design constraints associated with the nearly passive nature of RIS elements which cannot perform channel estimation directly, the opportunities offered by RIS for redefining the traditional notion of communication without producing new EM signals but by recycling existing radio waves, and the choice of using cost-effective material in realizing RIS to promote more sustainable wireless by design. These characteristics offer new opportunities for customizing the wireless. These characteristics pose new challenges in designing RIS-assisted networks/systems (e.g., communication, sensing, wireless charging, etc.), such as information transfer within the RIS-embedded environment, RIS configuration optimization with limited information, resource allocation and network optimization in such communication systems as discussed in [6], [7]

Objectives

Results

Conclusion