My research interests are mainly in the field of context awareness and its applicability to pervasive and ubiquitous computing.

I am interested in how context can be used to develop spontaneous ad hoc applications, for example, a video conferencing where some interesting questions impose themselves:

  what services should be provided to users in the conference?
  How to locate these services and how to invoke them?
  how presence of users in the meeting be projected to each other without violating privacy issues?

 

My research at MMARL group explores the interactions between mobile, pervasive computing, multi-agent systems and context awareness. I am currently developing an infrastructure that supports context to achieve better understanding and discovery of the environment where the spontaneous application is running.

The Project is called Agent-based Context-Aware Infrastructure (ACAI)

The ACAI system is an infrastructure for context based pervasive applications, a future prospective for what personal networks might look like, combining the traditional networking methods with the latest wireless ad hoc network technology and multiagent technology capabilities.

Specifically: I developed the following components to ease the development and management of context-aware system:

  1. Set of ontologies for modeling context and meta-context information.
  2. A multi-agent system for realizing the functional architecture of ACAI.
  3. An Inductive fuzzy reasoning approach for reasoning about uncertain context information.
  4. A context-based negotiation mechanism to support automated adaptability in context-aware applications.

 

The over-all architecture of ACAI is depicted in the following figure.

   

 untitled1.bmp (1120290 bytes)

In the Process of developing the ACAI infrastructure, some new protocols, besides existing ones, must be designed to have in consideration context acquisition, interpretation, aggregation and usage of context in providing better quality of service in the application and network layer

                  wpe5.jpg (39056 bytes)

For more information, you can read my publications.