Evaluation of 2D and 3D interest management techniques in the distributed virtual environment DiVE

We present DiVE, a networking framework for the rapid development of massively multiuser three-dimensional (3D) virtual world applications. The primary contribution of the paper is twofold. First, we investigate zone-based area of interest (AOI) techniques aimed at improving DiVE’s scalability. Specifically, we present an interest management technique that is based on dividing the world into two-dimensional (2D) zones and a ‘double’ (inner/outer) AOI concept. Here the novelty is the investigation of different movement patterns of client entities to determine the optimal AOI/zone ratio. The empirical evaluation of our approach shows promising results on four metrics: bandwidth, frames per second, CPU, and RAM. Second, we extend the AOI concept from 2D space to 3D space, which has not yet been addressed in the literature. Among others, 3D space suggests interesting applications for the safety of unmanned aerial vehicles. Besides the investigation of the optimal size of volumetric AOI, we also consider different forms of volumetric zone alignment. The secondary contribution consists in a technical description of the DiVE framework, which has already been used successfully in multiuser applications in the traffic and evacuation domains.