Informing Guessing Attacks on Publicly Performed Secrets Laura - - PowerPoint PPT Presentation

informing guessing attacks on publicly
SMART_READER_LITE
LIVE PREVIEW

Informing Guessing Attacks on Publicly Performed Secrets Laura - - PowerPoint PPT Presentation

Informing Guessing Attacks on Publicly Performed Secrets Laura South Mentors: Janne Lindqvist & Gradeigh Clark July 2017 Motivation Experimental setup Outline Simple patterns Complex patterns Next steps Motivation Research


slide-1
SLIDE 1

Informing Guessing Attacks on Publicly Performed Secrets

Laura South Mentors: Janne Lindqvist & Gradeigh Clark July 2017

slide-2
SLIDE 2

Outline

Motivation Experimental setup Simple patterns Complex patterns Next steps

slide-3
SLIDE 3

Motivation

  • Research question: can information about passwords be obtained by
  • bserving a person unlock a mobile device at a distance?
  • Similar efforts in recent research:
  • Focused on hand/finger observation at close distance where device is
  • bservable (Ye et al., 2017)
  • Used other methods (accelerometer) to obtain information from publicly

performed secrets (Owusu, Han, Das, Perrig, & Zhang, 2012)

slide-4
SLIDE 4

Experimental Setup

Step 1: Camera work

  • Two camera orientations

Side: Back:

  • Two tracking points
  • side orientation: elbow & wrist
  • back orientation: elbow & shoulder
slide-5
SLIDE 5

Experimental Setup

Step 2: Motion tracking

  • Software: Kinovea (open source video analysis)
slide-6
SLIDE 6

Experimental Setup

XML

Output from motion tracking software

Java

Extracts tracking info from XML, discards the rest

R

Creates plot using information from Java program

Step 3: Data visualization

slide-7
SLIDE 7

How to read a movement plot

Starting Point Gesture performed:

slide-8
SLIDE 8

Simple patterns

Pattern performed Direction

  • f wrist

motion Direction

  • f elbow

motion Up up-right down-right Down down-left up-left Left up-right down-right Right down-left up-left

slide-9
SLIDE 9

Simple patterns

  • Side orientation
  • Four diagonal movements (upleft

downright, upright, downleft)

slide-10
SLIDE 10

Moving on to more complex patterns

1 2 3 4 1 2 3 4 Pattern performed Direction of wrist movement Up up-right Down down-left Left up-right Right down-left

slide-11
SLIDE 11

Next steps

  • 3D depth sensing using Project Tango tablet or Kinect
  • Differentiate more clearly between “up” vs. “left” and “down” vs. “right"
  • Analyze data from back orientation
  • Expand dataset to include a more diverse group of subjects
  • Create movement classifier
slide-12
SLIDE 12

Acknowledgements

slide-13
SLIDE 13

References

  • 1. Owusu, E., Han, J., Das, S., Perrig, A., & Zhang, J. (2012, February).

ACCessory: password inference using accelerometers on

  • smartphones. In Proceedings of the Twelfth Workshop on Mobile

Computing Systems & Applications (p. 9). ACM.

  • 2. Ye, G., Tang, Z., Fang, D., Chen, X., Kim, K. I., Taylor, B., & Wang, Z.

(2017, January). Cracking Android pattern lock in five attempts. In The Network and Distributed System Security Symposium.