IoT-ID: Robust IoT Device Identification Based on Feature Drift Adaptation

Qi Chen, Yubo Song, Brendan Jennings, Fan Zhang, Bin Xiao, Shang Gao

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

1 Citation (Scopus)

Abstract

Internet of Things (IoT) devices deployed in publicly accessible locations increasingly encounter security threats from device replacement and impersonation attacks. Unfortunately, the limited memory and poor computing capability on such devices make solutions involving complex algorithms or enhanced authentication protocols untenable. To address this issue, device identification technologies based on traffic characteristics finger-printing have been proposed to prevent illegal device intrusion and impersonation. However, because of time-dependent distribution of traffic characteristics, these approaches often become less accurate over time. Meanwhile insufficient attention has been paid to the impact of possible changes on the accuracy of device identification. Therefore, we propose a novel feature selection method based on degree of feature drift and genetic algorithm to keep high accuracy and stability of device identification. The degree of feature drift - relevance of features through time and gain ratio are combined as a composite metric to filter out stable features. Furthermore, in order to perform equally well in device identification, we use the genetic algorithm to select the most discriminate feature subset. Experiments show that the accuracy of device recognition compared with other methods is increased from 86.4% to 94.5%, and the robustness of recognition is also improved.

Original languageEnglish
Title of host publication2021 IEEE Global Communications Conference, GLOBECOM 2021 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-6
ISBN (Electronic)9781728181042
DOIs
Publication statusPublished - Dec 2021
Event2021 IEEE Global Communications Conference, GLOBECOM 2021 - Madrid, Spain
Duration: 7 Dec 202111 Dec 2021

Publication series

Name2021 IEEE Global Communications Conference, GLOBECOM 2021 - Proceedings

Conference

Conference2021 IEEE Global Communications Conference, GLOBECOM 2021
Country/TerritorySpain
CityMadrid
Period7/12/2111/12/21

Keywords

  • device fingerprinting
  • feature selection
  • genetic algorithm
  • IoT device identification
  • machine learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality
  • Health Informatics

Fingerprint

Dive into the research topics of 'IoT-ID: Robust IoT Device Identification Based on Feature Drift Adaptation'. Together they form a unique fingerprint.

Cite this