Authenticating a Known User Through Behavioral Biometrics Using a Smartphone Accelerometer

Abstract

This thesis investigates the feasibility of authenticating a user through a behavioral biometric signature from smartphone accelerometer data. Using a Samsung Galaxy S7, acceleration in relation to the necessary equilibrium, postural state for a subject to orient a smartphone in order to read a headline article was measured and recorded by the MATLAB Mobile application. Twenty subjects1 known and 19 unknownwere used in the creation of a MATLAB machine-learning classifier. The classifier accurately distinguished an unknown subject from the known subject. Recommendations for future work include repeating the experiment with the latest smartphone devices as available, incorporating different sensors available to the MATLAB Mobile App, and introducing noise to spoof the known user.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 2017
Accession Number
AD1082655

Entities

People

  • Patrick W. Jones

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Autonomy
  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Authentication
  • Biometric Security
  • Biometrics
  • California
  • Computers
  • Coordinate Systems
  • Department Of Defense
  • Institutional Review Board
  • Machine Learning
  • Mobile Application Software
  • Mobile Devices
  • Mobile Phones
  • Security
  • Supervised Machine Learning
  • United States
  • United States Naval Academy
  • United States Special Operations Command

Fields of Study

  • Computer science

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Business Analytics
  • Image Processing and Computer Vision.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks