Pervasive Computing Framework development

1
Pervasive Computing Framework development

• Kartik Vishwanath
• Arvind S. Gautam
• Rahul Gupta
• Sachin Singh

2
What is a Pervasive Computing Framework ?

• Framework for developing Smart environments
• Framework for integrating devices and services
  into a cohesive unit.
• Framework for providing services based on user
  context.

3
Limitations of Existing Frameworks

• Pervasive computing systems developed till date
  are ad-hoc.
• The autonomous systems are isolated and have no
  interaction.
• No concept of global pervasiveness

4
Our Goals

• To create a generic, adaptable framework for
  pervasive computing
• To create a global framework i.e. Autonomous
  systems are not isolated
• Abstract devices as services and perform dynamic
  composition of these based on user context.

5
AS Head
Unit 1 (Service)
Unit 3
Device 1
Unit 2
Device 2
Device 3
6
AS HEAD 1
AS HEAD 3
AS HEAD 2
7
Our Solution

• Event Based autonomous system (AS)
• AS defined by a set of policies
• User identified by profile
• User Profile AS policies Services available
  to a user

8
.Our Solution

• Devices abstracted by services
• Semantic Services use user context to resolve
  atomic services
• AS Policies can restrict use of certain services
• Two layers of services
• Atomic Services
• Semantic Services

9
(No Transcript)
10
The Current Scenario

• The Hospital environment can be modeled into the
  system.
• Various devices (MRI machines etc.) must be
  abstracted
• The Knowledge Base developed by the KM group can
  drive Semantic Services

11
Study of Various Technologies

• One.World A framework that facilitates building
  of applications for pervasive computing.
• Developed by New York University
• http//www.cs.nyu.edu/rgrimm/one.world/

12
One.World Architecture
13
Components of One.World

• The design has the following components
• Virtual machine
• Support ad hoc composition of services
• Tuples (self-describing data)
• Simplify sharing
• Environments
• Act like address spaces, including protection
• Store persistent data
• Facilitate composition, check pointing, migration
• Events
• Make change explicit to the application

14
What we can use from One.World?

• Concept of dynamic and distributed resource
  discovery servers compared to JINI that has
  static discovery servers.
• Mechanisms for checkpointing and storing
  application state.
• Mechanism of Event passing.

15
Temporal Reasoning

• Rule-based reasoner for processing and reasoning
  about temporal information given in XSD encoding.
• Functions of reasoner like
• Parsing instants of time in XSD representation
• Constructs for manipulating default time zone
  used in process of normalizing non-UTC time to
  UTC time.
• Constructs for comparing temporal order (before,
  after, and equals) for time instants.
• Built inference support for high-level temporal
  reasoning.
• Constructs for comparing time instants and time
  intervals.

16
An Ontology of Time