Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation - Robust ML-based physical layer authentication using AoA

Aug 4, 2026

The paper investigates angle of arrival (AoA) as a feature for robust machine learning–based physical layer authentication (PLA), addressing the gap left by prior PLA work that mainly used channel frequency/impulse response or received signal strength. It analyzes impersonation attacks and proves that effective success requires highly restrictive attacker conditions related to location and hardware capabilities, enabling AoA to serve as a reliable authentication feature in many scenarios. It further develops lightweight, model-free ML approaches and validates them on experimental outdoor massive MIMO data.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Physical Layer Authentication in 6G/IoT

AoA as an Authentication Feature

Related Work on ML-based PLA

Contributions and Outline

UNIVERSITÀ POLITECNICA DELLE MARCHE Repository ISTITUZIONALE Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation This is a pre print version of the following article: Original Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation / Pham, Thuy M.; Senigagliesi, Linda; Baldi, Marco; P. , Fettweis Gerhard; Chorti, Arsenia. - ELETTRONICO. - (2023), pp. 13-18. ( IEEE Global Communications Conference (GLOBECOM) 2023 Kuala Lumpur (Malaysia) 4–8 December 2023) [10 . 1109/GLOBECOM54140 .2023. 10437915] . Availability: This version is available at: 11566/325840 since: 2024-01-08T16:20:31Z Publisher: Institute of Electrical and Electronics Engineers Inc. Published DOI:10.1109/GLOBECOM54140.2023.10437915 Terms of use: The terms and conditions for the reuse of this version of the manuscript are specified in the publishing policy. The use of copyrighted works requires the consent of the rights’ holder (author or publisher) . Works made available under a Creative Commons license or a Publisher's custom-made license can be used according to the terms and conditions contained therein. See editor’s website for further information and terms and conditions. This item was downloaded from IRIS Università Politecnica delle Marche ([https://iris.univpm.it](https://iris.univpm.it)) . When citing, please refer to the published version. (Article begins on next page) 10 March 2026 Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation This paper was downloaded from TechRxiv ([https://www.techrxiv.org](https://www.techrxiv.org)) . LICENSE CC BY 4.0 SUBMISSION DATE / POSTED DATE 12-10-2023 / 16-10-2023 CITATION Pham , Thuy M. ; Signialesi , Linda; Baldi , Marco; Fettweis , Gerhard P. ; Chorti , Arsenia (2023) . Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation. TechRxiv. Preprint. [https://doi.org/10.36227/techrxiv.24298384.v1](https://doi.org/10.36227/techrxiv.24298384.v1) DOI 10.36227/techrxiv.24298384.v1 Machine Learning-based Robust Physical Layer Authentication Using Angle of Arrival Estimation Thuy M. Pham⋆ , Linda Senigagliesi†, Marco Baldi†, Gerhard P. Fettweis⋆ , Arsenia Chorti⋆‡ ⋆Barkhausen Institut, Dresden, Germany, †Università Politecnica delle Marche, Ancona, Italy, ‡ETIS UMR 8051, CYU, ENSEA, CNRS, Cergy, France {minhthuy.pham, [gerhard.fettweis}@barkhauseninstitut.org](gerhard.fettweis}@barkhauseninstitut.org), {l.senigagliesi, [m.baldi}@univpm.it](m.baldi}@univpm.it) [arsenia.chorti@ensea.fr](arsenia.chorti@ensea.fr) Abstract—In this paper, we study the use of the angle of arrival (AoA) as a feature for performing robust, machine learning (ML) -based physical layer authentication (PLA). In fact, whereas most previous research on PLA relies on physical properties such as channel frequency/impulse response or received signal strength, the use of the AoA in this context has not yet been studied in depth as a means of providing resistance to impersonation (spoofing) attacks. In this study, we first prove that an effective impersonation attack on AoA-based PLA can only succeed under very stringent conditions on the attacker in terms of location and hardware capabilities, and thus, the AoA can in many scenarios be used as a robust feature for PLA. In addition, we exploit machine learning in our study to perform lightweight, model-free, intelligent PLA. We show the effectiveness of the proposed AoAbased PLA solutions by testing them on experimental outdoor massive multiple input multiple output data. Index Terms—Authentication, physical layer authentication, angle of arrival, impersonation, spoofing, machine learning. I. INTRODUCTION The massive deployment of Internet of things (IoT) devices with constrained resources in beyond fifth generation (B5G) networks poses significant security risks. Conventional upperlay

FAQs

Why is angle of arrival (AoA) considered for physical layer authentication in this study?

The study uses AoA as an identity feature for robust PLA, motivated by earlier PLA research relying mainly on channel impulse/frequency or received signal strength.

Under what conditions can impersonation attacks against AoA-based PLA succeed?

The paper shows that effective impersonation succeeds only under very stringent attacker constraints, especially regarding attacker location and hardware capabilities.

How does the paper implement lightweight, model-free machine learning for PLA?

It exploits machine learning to construct lightweight, model-free intelligent PLA solutions and evaluates them using experimental outdoor massive MIMO data.

Aug 4, 2026

Upload by Theodore