The User-Context Module provides the capabilities needed for Internet networks to become user-centric, sensing where and how the user connects, the available network, and so on.
In the beginning, the Internet as developed by DARPA was designed as a means of communication between two hosts wanting to talk to each other. Today this model is no longer viable. It’s clear that in future the vast “network of networks” must become user-centric. So the Internet must start today to view not only the interconnected network nodes but also the end user and his/her context as a single entity. This is what researchers at the National University of Singapore are advocating, proposing the integration of their "User-Context Module" into the Internet architecture of tomorrow. The stated aim: to eventually improve the user’s experience by adjusting networks and applications in line with his/her needs and resources, the terminal being used, physical location, and the available network: in short, according to the context.
A specific module to capture contextual information
The researchers looked to cognitive science, which deals with the way information is represented and processed by the brain. Human thought is sensitive to context and to its environment, which helps us to take the right decisions. In order to attain this type of intelligence, the ideal architecture must be able to separate out the host, the host address, the end-user name, and network data and peripherals, and be able to manipulate them all as a single entity. The suggested device – the User-Context Module – aims to integrate relevant contextual information.
Three interconnecting layers on top of the infrastructure
The Module is made up of three subsystems: the Context Interface subsystem, the Context Model subsystem, and the Control subsystem. The first interacts directly with the end user and his/her environment. It collects information relevant to the context (preferences, physical location, identities, etc). The second collects information on the Context Model, abstracting data on user status and network behaviour. Finally, the Control subsystem interacts directly with the network layers and the infrastructure to automatically adjust the protocols and parameters so as to optimise network performance and thus provide the user with a better overall experience.