RIS phase optimization for near-field 5G positioning: ML-enhanced CRLB minimization

Abstract

This article addresses near-field localization using Reconfigurable Intelligent Surfaces (RIS) in 5G systems, where Line-of-Sight (LOS) between the base station and the user is obstructed. We propose a RIS phase optimization method based on the minimization of the Cramér-Rao Lower Bound (CRLB). This minimization itself is computationally costly, for which a data-driven method is employed with remarkable computational savings and positioning performance. The main contributions of this work are: (1) the application of machine learning (ML) to enhance CRLB minimization for RIS phase optimization; (2) an overview on RIS phases preprocessing methods to enhance deep neural networks training for the task; and (3) an end-to-end simulation of the positioning task with the presented method, showing a computational improvement without compromising positioning accuracy.

The UPC authors are within the Signal Processing and Communications Group at UPC recognized as a consolidated research group by the Generalitat de Catalunya through 2021 SGR 01033. This publication is part of the project ROUTE56 with grant PID2019-104945GB-I00 funded by MCIN/AEI/ 10.13039/501100011033 and the project 6-SENSES with grant PID2022-138648OB-I00 funded by MCIN/AEI/ 10.13039/501100011033 and by ERDF A way of making Europe. This work has been partially supported by the National Science Foundation under Awards ECCS-1845833 and CCF-2326559.

Peer Reviewed

Postprint (author's final draft)