A trust assurance technique for Internet of things based on human behavior compliance

Marco Anisetti, Claudio Agostino Ardagna, Ernesto Damiani, Alessandro Sala

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

The advent of the Internet of things (IoT) has radically changed the way in which computations and communications are carried out. People are just becoming another component of IoT environments, and in turn, IoT environments are becoming a mixture of platforms, software, services, things, and people. The price we pay for such dynamic and powerful environment is an intrinsic uncertainty and low trustworthiness due to its opaque perimeter, the multitude of different data sources with unknown providers, and uncertain responsibilities. Trustworthiness of observables collected by smart devices (from minuscle sensors to bigger machines) is fundamental to build a chain of trust on a decision process taken according to these observables. Some assurance solutions evaluate the quality of collected data, although they are difficult to apply in IoT environments for performance and cost reasons. In this paper, we take a different approach and put forward the idea that, in many cases, the behavior of people owning smart devices can contribute to the evaluation of the trustworthiness of collected data and, in turn, of the whole decision process. We therefore define an assurance methodology based on data analytics evaluating the compliance of people to behavioral policies. The more people behavior is compliant, the higher the trustworthiness of data collected through their smart devices.

Original languageBritish English
Article numbere5355
JournalConcurrency and Computation: Practice and Experience
Volume33
Issue number4
DOIs
StatePublished - 25 Feb 2021

Keywords

  • assurance
  • behavioral analysis
  • big data
  • Internet of Things

Fingerprint

Dive into the research topics of 'A trust assurance technique for Internet of things based on human behavior compliance'. Together they form a unique fingerprint.

Cite this