A Motion-aware Data Transfers Scheduling for Distributed Virtual Walkthrough Applications

distributed virtual walkthrough, next location prediction, motion function, Markov chain, prefetching, virtual environments

Data transfers scheduling is an important part of almost all distributed virtual walkthrough (DVW) applications. Its main purpose is to preserve data transfer efficiency and render quality during scene exploration. The most limiting factors here are network restrictions such as low bandwidth and high latency. Current scheduling algorithms use multi-resolution data representation, priority determination and data prefetching algorithms to minimize these restrictions. Advanced priority determination and data prefetching methods for DVW applications use mathematic description of motion to predict next position of each individual user. These methods depend on the recent motion of a user so that they can accurately predict only near locations. In the case of sudden but regular changes in user motion direction (road networks) or fast moving user, these algorithms are not sufficient to predict future position with required accuracy and at required distances. In this paper we propose a systematic solution to scheduling of data transfer for DVW applications which uses next location prediction methods to compute download priority or additionally prefetch rendered data in advance. Experiments show that compared to motion functions the proposed scheduling scheme can increase data transfer efficiency and rendered image quality during scene exploration.