Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

6G Fresnel Spot Beamfocusing using Large-Scale Metasurfaces: A Distributed DRL-Based Approach

Mehdi Monemi, Mohammad Amir Fallah, Mehdi Rasti, Matti Latva-Aho

TL;DR

A modular highly scalable structure composed of multiple sub-arrays, each equipped with a TD3 deep-reinforcement-learning (DRL) method enables collaborative optimization of the radiated power at the DFP, significantly reducing computational complexity while enhancing learning speed.

Abstract

In this paper, we introduce the concept of spot beamfocusing (SBF) in the Fresnel zone through extremely large-scale programmable metasurfaces (ELPMs) as a key enabling technology for 6G networks. A smart SBF scheme aims to adaptively concentrate the aperture's radiating power exactly at a desired focal point (DFP) in the 3D space utilizing some Machine Learning (ML) method. This offers numerous advantages for next-generation networks including efficient wireless power transfer (WPT), interference mitigation, reduced RF pollution, and improved information security. SBF necessitates ELPMs with precise channel state information (CSI) for all ELPM elements. However, obtaining exact CSI for ELPMs is not feasible in all environments; we alleviate this by proposing an adaptive novel CSI-independent ML scheme based on the TD3 deep-reinforcement-learning (DRL) method. While the proposed ML-based scheme is well-suited for relatively small-size arrays, the computational complexity is unaffordable for ELPMs. To overcome this limitation, we introduce a modular highly scalable structure composed of multiple sub-arrays, each equipped with a TD3-DRL optimizer. This setup enables collaborative optimization of the radiated power at the DFP, significantly reducing computational complexity while enhancing learning speed. The proposed structures benefits in terms of 3D spot-like power distribution, convergence rate, and scalability are validated through simulation results.

6G Fresnel Spot Beamfocusing using Large-Scale Metasurfaces: A Distributed DRL-Based Approach

TL;DR

A modular highly scalable structure composed of multiple sub-arrays, each equipped with a TD3 deep-reinforcement-learning (DRL) method enables collaborative optimization of the radiated power at the DFP, significantly reducing computational complexity while enhancing learning speed.

Abstract

In this paper, we introduce the concept of spot beamfocusing (SBF) in the Fresnel zone through extremely large-scale programmable metasurfaces (ELPMs) as a key enabling technology for 6G networks. A smart SBF scheme aims to adaptively concentrate the aperture's radiating power exactly at a desired focal point (DFP) in the 3D space utilizing some Machine Learning (ML) method. This offers numerous advantages for next-generation networks including efficient wireless power transfer (WPT), interference mitigation, reduced RF pollution, and improved information security. SBF necessitates ELPMs with precise channel state information (CSI) for all ELPM elements. However, obtaining exact CSI for ELPMs is not feasible in all environments; we alleviate this by proposing an adaptive novel CSI-independent ML scheme based on the TD3 deep-reinforcement-learning (DRL) method. While the proposed ML-based scheme is well-suited for relatively small-size arrays, the computational complexity is unaffordable for ELPMs. To overcome this limitation, we introduce a modular highly scalable structure composed of multiple sub-arrays, each equipped with a TD3-DRL optimizer. This setup enables collaborative optimization of the radiated power at the DFP, significantly reducing computational complexity while enhancing learning speed. The proposed structures benefits in terms of 3D spot-like power distribution, convergence rate, and scalability are validated through simulation results.
Paper Structure (20 sections, 1 theorem, 28 equations, 13 figures, 1 table, 3 algorithms)

This paper contains 20 sections, 1 theorem, 28 equations, 13 figures, 1 table, 3 algorithms.

Table of Contents

  1. Introduction
  2. Introduction
  3. Background work
  4. Motivation and Contributions
  5. System Model and Problem Formulation
  6. System Model
  7. Fresnel Zone Considerations
  8. Problem Formulation
  9. Proposed Solution Scheme
  10. The overall structure of the proposed system
  11. Sub-array beamfocusing through TD3-DRL scheme
  12. Environment
  13. Actor
  14. Critic
  15. Agent
  16. ...and 5 more sections

Key Result

Theorem 1

Let $\boldsymbol{w}_1^*$ and $\boldsymbol{w}_2^*$ be the solutions to P1 and P2 respectively. The optimal SBF problem P1 and WPT problem P2 for ELPMs are equivalent for the focal reference plane, in the sense that $R(\boldsymbol{w}_2^*,\boldsymbol{r}^U,\eta)= R(\boldsymbol{w}_1^*,\boldsymbol{r}^U,

Figures (13)

  • Figure 1: System model of the ELPM SBF system
  • Figure 2: Different scenarios for beamfocusing wherein an aperture is located on the $xz$ plane and the focal point is located on the $xy$ plane in front of the aperture: (a) DFP located in the non-radiating near-field region. (b) DFP located in the far-field region. (c) DFP located in the Fresnel region using a small-scale PM. (d) SBF at the DFP realized in the Fresnel region using ELPM.
  • Figure 3: Two BFRs corresponding to two different values of $\eta$. The yellow and red circles respectively have BFRs $R_1$ and $R_2$, containing $\eta_1=90$% and $\eta_2=$50% of the total power in the reference plane.
  • Figure 4: Hardware structure of the proposed modular SBF system for a sample case consisting of $M=9$ modules.
  • Figure 5: TD3-DRL based software structure for the proposed modular Fresnel zone SBF system showing the process of training sub-array modules and the determination of the final beamforming vector after convergence.
  • ...and 8 more figures

Theorems & Definitions (7)

  • Remark 1
  • Remark 2
  • Theorem 1
  • proof
  • Remark 3
  • Remark 4
  • Remark 5