An Adaptive Context-Aware Transaction Model for Mobile and Ubiquitous Computing

Abstract

Transaction management for mobile and ubiquitous computing (MUC)aims at providing mobile users with reliable and transparent services anytime anywhere. Traditional mobile transaction models built on client-proxy-server architecture cannot make this vision a reality because (1) in these models, base stations (proxy) are the prerequisite for mobile hosts (client) to connect with databases (server), and 2)few models consider context-based transaction management. In this paper, we propose a new network architecture for MUC transactions, with the goal that people can get online network access and transaction even while moving around; and design a context-aware transaction model and a context-driven coordination algorithm adaptive to dynamically changing MUC transaction context. The simulation results have demonstrated that our model and algorithm can significantly improve the successful ratio of MUC transactions.