Mobile computing equipment such as laptop computers, hand-held devices,
and perhaps even wearable computers will soon be in everyday use by
everyone, anywhere and anytime. Users are no longer tied to their
offices and homes. They will carry a computer with them, and use it
while on the move. This introduces a completely new concept to be taken
into account by software developers --- the location of the user.
Supporting nomadic users raises great challenges; communication over a
wireless connection must be optimized, and new kinds of applications
should be developed. Because of the very dynamic wireless environment,
applications must be capable of adapting to the circumstances and making
decisions autonomously. Thus, use of intelligent agents to support
nomadic users is a natural approach. In this paper, we tackle the
problem of predicting the future location of a user, which is often the
base for the intelligent behavior of nomadic application agents. We
present a method for learning the regular routes of a user, and show how
movement predictions can be used to predict the future quality of a
wireless connection.
Publishing Information
In FDPW'99.
Electronic Versions
Paper (PDF) (may not display properly inside
Netscape)