Deducing User Presence from Inter-Message Intervals in Home Automation Systems - ICT Systems Security and Privacy Protection
Conference Papers Year : 2016

Deducing User Presence from Inter-Message Intervals in Home Automation Systems

Frederik Möllers
  • Function : Author
  • PersonId : 989272
Christoph Sorge
  • Function : Author
  • PersonId : 989273

Abstract

Privacy in Home Automation Systems is a topic of increasing importance, as the number of installed systems constantly grows. In this paper we investigate the ability of an outside observer to link sets of message timestamps together to predict user presence and absence. The question we try to answer is: If attacker Eve has captured 1 hour of traffic from victim Alice’s HAS and knows whether Alice was present at that time, can Eve deduce Alice’s state by capturing another hour of traffic? We apply different statistical tests and show that in certain situations, the attacker can infer the user’s presence state with absolute confidence.
Fichier principal
Vignette du fichier
421518_1_En_25_Chapter.pdf (1.1 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01369569 , version 1 (21-09-2016)

Licence

Identifiers

Cite

Frederik Möllers, Christoph Sorge. Deducing User Presence from Inter-Message Intervals in Home Automation Systems. 31st IFIP International Information Security and Privacy Conference (SEC), May 2016, Ghent, Belgium. pp.369-383, ⟨10.1007/978-3-319-33630-5_25⟩. ⟨hal-01369569⟩
106 View
95 Download

Altmetric

Share

More