Graduate Thesis Or Dissertation
 

HiNoVa: A Novel Open-Set Detection Method for Automating RF Device Authentication

Public Deposited

Contenu téléchargeable

Télécharger le fichier PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/rx913z59p

Descriptions

Attribute NameValues
Creator
Abstract
  • New capabilities in wireless network security are now possible through deep learning, which can identify and leverage patterns in radio frequency (RF) data. One area of deep learning, known as open set detection, is focused on detecting data instances from new devices encountered during deployment that were not previously seen in the training set. Open-set detection is a promising approach for detecting unauthorized devices, where these devices correspond to new devices. Although open set detection has received considerable attention in the literature, most of the past work has been applied to images and is intended to be applied to independent and identically distributed data instances. In contrast, RF signals present a unique set of challenges as the data form a time series with many frequencies embedded within it. I introduce a novel open-set detection approach based on the activation patterns of the hidden states within a Convolutional Neural Network Long Short-Term Memory (CNN+LSTM) model. My approach greatly improves the Area Under the Precision-Recall Curve (AUPRC) on LoRa, Wireless-WiFi, and Wired-WiFi datasets.
Contributor
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Déclaration de droits
Publisher
Peer Reviewed
Language
Embargo reason
  • Pending Publication
Embargo date range
  • 2023-03-24 to 2024-04-25

Des relations

Parents:

This work has no parents.

Dans Collection:

Articles

La vignette Titre Date de téléchargement Visibilité actes
file details PuppoLukeA2023.pdf 2023-03-28 Public Télécharger