vix.ing · top · new · best · stats · spec

Deep Learning-based Phase Reconfiguration for Intelligent Reflecting\n Surfaces

2020/09/29 by Özgecan Özdoğan, Emil Björnson, Özdogan, Özgecan +1 · 3 citations
Engineering · Materials Science · #Advanced Antenna and Metasurface Technologies #Advanced Wireless Communication Technologies #FOS: Computer and information sciences #FOS: Electrical engineering #Information Theory (cs.IT) #Machine Learning (cs.LG) #Metamaterials and Metasurfaces Applications #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2009.13988

openalex publication_date 2020/09/29 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28

Abstract

Intelligent reflecting surfaces (IRSs), consisting of reconfigurable\nmetamaterials, have recently attracted attention as a promising cost-effective\ntechnology that can bring new features to wireless communications. These\nsurfaces can be used to partially control the propagation environment and can\npotentially provide a power gain that is proportional to the square of the\nnumber of IRS elements when configured in a proper way. However, the\nconfiguration of the local phase matrix at the IRSs can be quite a challenging\ntask since they are purposely designed to not have any active components,\ntherefore, they are not able to process any pilot signal. In addition, a large\nnumber of elements at the IRS may create a huge training overhead. In this\npaper, we present a deep learning (DL) approach for phase reconfiguration at an\nIRS in order to learn and make use of the local propagation environment. The\nproposed method uses the received pilot signals reflected through the IRS to\ntrain the deep feedforward network. The performance of the proposed approach is\nevaluated and the numerical results are presented.\n

Cited by

Related