12-05-2011, 03:20 PM
[attachment=13660]
[attachment=13661]
INTRODUCTION
PERVASIVE ENVIRONMENT
Pervasive computing is a rapidly developing area of Information and Communications Technology (ICT).
Pervasive computing involves three converging areas of
ICT:
Computing (‘devices’),
Communication (‘connectivity’) and
User interfaces.
VIRTUAL WALL
As sensor-rich pervasive environments become more common, users’ privacy will be at Increased risk. Sensors can record a user’s activities and personal information such as heart rate, body temperature, and even conversations. Users may unwittingly leave “Digital footprints” that can threaten their privacy
DIGITAL FOOTPRINTS
A Digital Footprint is the data trace or trail left by someone activity in a digital environment.
Digital Footprints are invisible but provide data on what you have done, where the person has been, how long they stayed, both via the web and by location (geography), how often, the route or routine and increasingly who with, your social crowd.
LEVEL OF TRANSARENCY FOR VIRTUAL WALL
TRANSPARENT
OPAQUE
TRANSLUCENT
TRANSPARENT VIRTUAL WALL
Alice and Bob are aware of their physical privacy within the meeting room, but sensors are leaking personal footprints to the pervasive environment.
TRANSLUCENT VIRTUAL WALL
Alice and Bob deploy a translucent virtual wall to prevent unwanted disclosure of personal footprints.
ARCHITECTURE
Our system consist of :-
Context server.
Client.
Sensing infrastructure.
Users instantiate walls by contacting the context server, which then use the virtual walls
to regulate access to footprints.
SECURITY ASSUMPTIONS
Trusted context server.
Secure location claim.
Sensor security.
VIRTUAL WALL MODEL
PLACES
FOOTPRINTS
QUERIES
VIRTUAL WALL
VIRTUAL WALL INTERACTION
Two virtual walls conflict if there exists a queries for whom two or more different transparencies apply for access to a particular digital footprint.
These conflicts can occur
between different walls owned by the same user.
by an opaque wall of one user that contradicts the transparent or translucent virtual walls of other users.
between different walls for different owners for a group-owned personal footprint.
LIMITATION OF MODEL
Virtual wall control all footprints within the personal and general categories uniformly. Users may want finer-grained control over some footprints and specialized mechanisms could be used in conjunction with virtual walls. We leave such a hybrid approach to future work.
Our model does not address multiple queries.
FUTURE WORK
The following challenges and opportunities for future work:-
Creating walls: It may be cumbersome for users to constantly think about deploying virtual walls for every place they visit.
Group ownership: A group of users may want to create a shared virtual wall to control footprints related to the group
User disruption :Opaque walls, and possibly group walls in the future, require input from and feedback to users.
Mobile places: One can envision mobile places, such as a bus enriched with sensors.For instance, the bus may be parked inside a building and so be affected by virtual walls around that building.We would like to explore the semantics of virtual walls for mobile places, and how they interact with static places.
Data perturbation: It may be desirable to perturb footprints when using translucent walls.
Deception :User studies in location privacy have identified the need for deception,where users can lie about their location
CONCLUSION
Virtual walls control the spread of personal and general contextual data, or “footprints,” and offer privacy analogous to that afforded by physical walls. . We formalized the semantics of access control using virtual walls, and evaluated our model through a user study. Our results indicate that the model is easy to understand and easy to use . Privacy in sensor-rich environments is an important problem that might affect the deployment of such environments. Based on our results, we believe that virtual walls are a promising metaphor for specifying usable privacy policies in pervasive environments.