Towards an Interest Management Scheme for Peer-based Virtual Environments

Abstract

A fundamental task in peer-to-peer-based Massively Multiuser Virtual Environments is providing all peers with a consistent view of the environment. To do so, state changes must be propagated to peers. To limit the resulting network traffic, existing approaches often restrict the distribution of a state change to peers for which it is relevant. The identification of such peers is known as Interest Management. Typically Interest Management is done based on so-called Areas of Interest. An Area of Interest is a spatial area around a user's avatar. If a change occurs inside this area, it is relevant and must be reported to the corresponding peer. We propose to extend this approach with so-called Areas of Effect. An Area of Effect specifies the spatial area of the virtual environment that is directly influenced by a given state change. A state change is propagated to all peers whose Areas of Interest intersect with the change's Area of Effect. This allows us to model complex state changes with arbitrary and possibly dynamic influence areas. They may even affect multiple areas at once. In this paper we describe our approach and give an overview on the current state of it's implementation.