Video-streaming for fast moving users in 3G mobile networks

Abstract

The emergence of third-generation (3G) mobile networks offers new opportunities for the effective delivery of data with rich content including multimedia messaging and video-streaming. Provided that streaming services have proved highly successful over stationary networks in the past, we anticipate that the same trend will soon take place in 3G networks. Although mobile operators currently make available pertinent services, the available resources of the underlying networks for the delivery of rich data remain inherently constrained. At this stage and in light of large numbers of users moving fast across cells, 3G networks may not be able to warrant the needed quality-of-service requirements. The support for streaming services necessitates the presence of content or media servers properly placed over the 3G network; such servers essentially become the source for streaming applications. Evidently, a centralized approach in organizing streaming content might lead to highly congested media-nodes which in presence of moving users will certainly yield increased response times and jitter to user requests. In this paper, we propose a workaround that enables 3G networks to offer uninterrupted video-streaming services in the presence of a large number of users moving in high-speed. At the same time, we offer a distributed organization for the network's mediaservers to better handle over-utilization.

Headlight prefetching for mobile media streaming

“…The V3 architecture proposed by Guo et al for live video streaming is essentially a P2P streaming architecture for moving vehicles [6]. Kyriakidou et al discuss streaming architecture, content distribution and rate adaptation algorithms for video streaming to fast moving users in 3G networks [9]. Zhai et al propose buffer management methods that employ statistical analysis to predict the trend of user movement among cells [16].…”