MoGATU: Data Management in Pervasive Computing Environments

September 1, 2001 - August 1, 2006

The past few years have seen significant work in mobile data management, typically based on the client/proxy/server model. Mobile/wireless devices are treated as clients that are data consumers only, while data sources are on servers that typically reside on the wired network. With the advent of pervasive computing environments, an alternative scenario arises where mobile devices gather and exchange data from not just wired sources, but also from their ethereal environment and one another. This is accomplished using ad-hoc connectivity engendered by Bluetooth like systems. The traditional client/server model is no longer appropriate to represent the new scenario; instead, mobile devices should be treated as peers. The mobile devices become both data consumers and providers, spontaneously interacting with other devices in their vicinity to pursue their individual and collective goals. The data-intensive, pervasive computing model for mobile ad-hoc networks poses new challenges for data management among these devices. These challenges include the spatio-temporal variation of data and data source availability, lack of a global catalog & schema, no guarantee of reconnection among peers, no guarantee of collaboration among peers, and the issues of commits and aborts due to the serendipitous nature of the environment.

In this project, we are solving the problem of data management in pervasive computing environments facing the above challenges. Our solution broadly consists of two parts: treating the devices as semi autonomous entities guided in their interactions by profiles & context, and the design of a contract-based transaction model. The profile is grounded in a semantically rich language for representing information about users, devices, and data objects each described in terms of "beliefs", "desires", and "intentions" - a model which has been explored in multi-agent interactions. We introduce data-based routing algorithms and semantic-based data caching and replication algorithms enabling mobile devices to utilize their data-intensive vicinities. We also introduce novel interactions autonomously initiated by each device while trying to pursue its user's "intentions" and "desires" encoded in the user profiles. We are developing transaction models based on contract net principles for peer-to-peer interaction and will demonstrate our findings and solutions by creating a lightweight peer-to-peer data management architecture in pervasive environments. We also investigate cross layer security issues, integrating data level trust with network level observations

Tags: ad-hoc networking, bluetooth, data management, discovery, ontology, pervasive computing, security, semantic web, wlan

OWL Tweet