Isaac Long

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2014-95

May 16, 2014

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-95.pdf

In this project, we explore two forms of user authentication, alternative to the alpha-numeric password, through utilization of various hardware components in a modern smartphone. These components act as input channels for biometric and geolocative data, which can be used to validate the identity of a user through gesture-based authentication (via the accelerometer and orientation sensor) and location-based authentication (via the WIFI sensor). We develop an Android application that incorporates these two authentication methods. We implement two different gesture-based authentication algorithms adapted from existing work done in this field of security research. We find that one algorithm is more convenient to train and implement than the other algorithm, but also performs more poorly in terms of accuracy compared to the other algorithm. We also implement a location-based authentication algorithm based on scanning the WIFI access points around the user’s current location. We introduce thresholds into the algorithm to provide tolerance and robustness against inconsistent WIFI access points, and these thresholds show improvement in the algorithm’s performance.

Advisors: Laurent El Ghaoui and Donald Wroblewski

---

BibTeX citation:

@mastersthesis{Long:EECS-2014-95,
    Author= {Long, Isaac},
    Title= {Leveraging Smartphone Hardware Capabilities for Alternative Authentication},
    School= {EECS Department, University of California, Berkeley},
    Year= {2014},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-95.html},
    Number= {UCB/EECS-2014-95},
    Abstract= {In this project, we explore two forms of user authentication, alternative to the alpha-numeric password, through utilization of various hardware components in a modern smartphone. These components act as input channels for biometric and geolocative data, which can be used to validate the identity of a user through gesture-based authentication (via the accelerometer and orientation sensor) and location-based authentication (via the WIFI sensor). We develop an Android application that incorporates these two authentication methods. We implement two different gesture-based authentication algorithms adapted from existing work done in this field of security research. We find that one algorithm is more convenient to train and implement than the other algorithm, but also performs more poorly in terms of accuracy compared to the other algorithm. We also implement a location-based authentication algorithm based on scanning the WIFI access points around the user’s current location. We introduce thresholds into the algorithm to provide tolerance and robustness against inconsistent WIFI access points, and these thresholds show improvement in the algorithm’s performance.},
}

EndNote citation:

%0 Thesis
%A Long, Isaac 
%T Leveraging Smartphone Hardware Capabilities for Alternative Authentication
%I EECS Department, University of California, Berkeley
%D 2014
%8 May 16
%@ UCB/EECS-2014-95
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-95.html
%F Long:EECS-2014-95