Towards context-aware internet services with unselfish clients

Abstract

The rapid advancement in context-aware computing techniques greatly facilitates capturing the context information of the Internet clients, which can be utilized by the Internet services and applications to manage different network resources. Based on the built context-aware system and the deduced highly abstract context information, we propose a resource distribution framework that incentivizes context sharing and moderate competition among the selfish but rational Internet clients. Under the proposed framework, the Internet client, which provides its both negative and positive context, can be assigned to the prioritized class and accordingly receive more resources from the resource owner. Meanwhile, all the clients are motivated by the framework to compete moderately and the aggressive ones are penalized by receiving fewer resources. The Web system exemplar is used to aid understanding of our motivation. We further model the resource distribution process as a non-cooperative game and accordingly provide the theoretical insight of the proposed framework.