Memory-Enhanced Dynamic Evolutionary Control of Reconfigurable Intelligent Surfaces

TL;DR

This paper introduces a memory-enhanced evolutionary control method for reconfigurable intelligent surfaces that optimizes communication throughput without needing detailed channel measurements, leveraging environment correlations.

Contribution

It presents a novel evolutionary control algorithm with memory for RISs, enabling efficient, adaptive optimization in complex communication environments without extensive channel knowledge.

Findings

Effective maximization of worst-case throughput

No need for separate channel matrix measurements

Demonstrated potential through numerical examples

Abstract

An innovative evolutionary method for the dynamic control of reconfigurable intelligent surfaces (RISs) is proposed. It leverages, on the one hand, on the exploration capabilities of evolutionary strategies and their effectiveness in dealing with large-scale discrete optimization problems and, on the other hand, on the implementation of memory-enhanced search mechanisms to exploit the time/space correlation of communication environments. Without modifying the base station (BS) beamforming strategy and using an accurate description of the meta-atom response to faithfully account for the micro-scale EM interactions, the RIS control (RISC) algorithm maximizes the worst-case throughput across all users without requiring that the Green's partial matrices, from the BS to the RIS and from the RIS to the users, be (separately) known/measured. Representative numerical examples are reported to illustrate the features and to assess the potentialities of the proposed approach for the RISC.