Tuyêt Trâm DANG NGOC's research

In a pervasive computing world, billions of devices are spread everywhere. All these devices can be seen as nodes of a giant, dynamical, unconnected directed graph

[Moving Graph]
  • a Graph, because there are nodes (data sources and terminals) connected by links (network wire or wireless links).
  • Directed because transmission between nodes can be assymetric (some sensor devices just emit data).
  • dynamical, because of nodes that appear and disappear in the graph as devices power on/off, or wireless mobile devices move out of reachability.
  • Unconnected as a consequence of the previous point, some nodes or part of the graph cannot be reachable.
  • and finally Giant, because the graph include the whole Internet and devices in the world.

This graph is composed of Links and Nodes, where:

[devices] Nodes of the graph belong to one of these three main categories:
  • Data Sources that provide data to other nodes of the graph. These are DBMS, web sites, file systems, sensors, etc.
  • Client Application that retrieve data from ther nodes of the graph. These are GUI, web application, terminals, etc.
  • some nodes are both client and source, and are called proxies. They take data from one our several nodes (sources or proxies), process the data and forward the result to other nodes of the graph (client or proxies). All middleware (mediator, etc.) match this category.

Links of the graph compose the graph infrastructure. There are several type of links:

  • permanents networks and ad-hoc networks.
  • wire or wireless networks.
  • networks based on TCP/IP or not.

The issue is to integrate this pervasive vision of computing devices in order to allow any user anywhere in the world to query anything from anywhere in the world.

Ie, how to organize the graph, so a user with any type of profile can connect on any node of the graph and access to other data that are dissiminated on nodes of the graph.

This objective leads to four sub-problems

[?]
  • Sources heterogeneity how to handle different type of sources having different models and types ?
  • Sources description how to locate data answering to the request and how to describe sources and their data. Finaly, fow to evaluate the query through the graph ?
  • Client heterogeneity how to adapt the client query depending on the access permission, the user preferences and the terminal capabilities ?
  • Data integration how to integrate data coming from different locations in order to evaluate a request ?

[Travaux] Links on my work
  • Data Integration
  • XQuery Evaluation
  • Mobile Agent Platform
  • Source Description
[Publications] My Publications
[PADAWAN] The PADAWAN Project
[Related] Links on Related Works
  • Data Integration
  • XQuery Evaluation
  • Mobile Agent Platform
  • Source Description
dntt@free.fr
Back to DNTT's HomePage