Trajectory Prediction for Routing Optimization in Pub/Sub Systems

In a Pub/Sub system, if the clients are permitted to move randomly, their mobility will cause their access brokers to be changed, which has brought in problems in dissemination of subscriptions and distribution of matching results. We claim that channelization-based routing algorithms can be applied to the Pub/Sub systems which support wireless clients. In this paper, we discuss the related issues of subscription and event distribution in a channelization-based Pub/Sub system. We also propose an optimization algorithm based on client trajectory prediction, which is proved to be a good solution for the problems brought by mobile clients and also can improve quality of service for mobile subscribers.