As much as ubiquitous computing systems are already claimed to exist in the real world, further development of these systems still pose challenges to computer science that are still quite beyond the state of the art. Two challenges standout in particular: the complexity of next-generation ubiquitous computing systems, and their inherent scalability issues. This paper aims to establish that agent-based modelling provides a powerful tool in tackling these issues. As an example of a practical solution, readily available, this paper highlights the distributed agent-based simulation infrastructure PDES-MAS as particularly suited for the task. Using the PDES-MAS infrastructure, designers, developers, and builders of next-generation ubiquitous computing systems can, through an iterative agent-based simulation process, gain the required knowledge and information about these systems, without having precede to deployment of the system itself.
Eleni ManginaJavier CarbóJosé M. Molina
Jian WuNoel N. SchulzWenzhong Gao