I think, therefore I am: usability and security of authentication using brainwaves

From Brede Wiki
Jump to: navigation, search
Conference paper (help)
I think, therefore I am: usability and security of authentication using brainwaves
Authors: John Chuang, Hamilton Nguyen, Charles Wang, Benjamin Johnson
Citation: Financial Cryptography and Data Security 7862 in Lecture Notes in Computer Science : 1-16. 2013
Editors: Andrew A. Adams, Michael Brenner, Matthew Smith
Publisher: Springer, Berlin Heidelberg
Meeting: FC 2013 Workshops
Database(s): Google Scholar cites
DOI: 10.1007/978-3-642-41320-9_1.
Link(s): http://www.kisc.meiji.ac.jp/~ethicj/USEC13/submissions/usec13_submission_06.pdf
Search
Web: DuckDuckGo Bing Google Yahoo!Google PDF
Article: Google Scholar PubMed
Restricted: DTU Digital Library
Services
Format: BibTeX

I think, therefore I am: usability and security of authentication using brainwaves is study of the subject-specific patterns of electroencephalography with a low-cost Neurosky MindSet device.

They find that among 15 subject they are able to identify subjects with a high accuracy based on data from only a single electrode.

Contents

[edit] Subjects

Subject group #1 (help)
Subjects
Description: UC Berkeley undergraduate or graduate students
Subjects/♂/♀: 15 /  /
Age: (–)
Nationality: define nationality
Approval: Institutional Review Board
Databases:

Group 1 of 15 subjects were included in the study.

The study on the human subjects was approved by the Institutional Review Board.


[edit] Method

  • Electroencephalography with the Neurosky Mindset electrode, corresponding to electrode Fp1.
  • 15 subjects, seven task, five times per session, two sessions per subject. A total of 15 x 7 x 5 x 2 = 1050 data set(?). Each subject was recorded for 40-50 minutes sessions on separate days.
  • Only data from the alpha wave band (8-12 Hz) and beta wave band (12-30).
  • Computation of the median magnitude across time for each frequency band.
  • For each subject: a vector with two values(?)
  • Comparing the features among subjects by the cosine similarity

[edit] Tasks

  1. "Breathing": Closed eyes
  2. Simulated finger movement
  3. "Sports task"
  4. ...

[edit] Results

  1. Single-channel EEG exhibit subject-specific patterns.
  2. Single-channel EEG biometrics ("authentication") is "just as accurate as multi-channel EEG" ("authentication").
  3. The signal is not affected by mental task, e.g., "personal pass-thought".

[edit] Related papers

  1. Electroencephalogram signals from imagined activities: a novel biometric identifier for a small population
  2. Low-cost electroencephalogram (eeg) based authentication
  3. My thoughts are not your thoughts
  4. Person authentication using brainwaves (eeg) and maximum a posteriori model adaptation
  5. Person identification based on parametric processing of the EEG
  6. Person identification from the eeg using nonlinear signal classification
  7. Two-stage biometric authentication method using thought activity brain waves
Personal tools