Network-assisted DASH by Utilizing Local Caches at Network Edge

In this paper, a solution for Dynamic Adaptive Streaming over HTTP(DASH), which utilizes local video caches available at the edge of the network is presented. The proposed solution uses Server and Network Assisted DASH(SAND) and edge computing solutions to provide an optimal Quality of Experience(QoE) for the end-user by utilizing locally cached content as much as possible. Utilization of cached content available at the edge of network reduces redundant traffic in the backhaul of access network and is useful especially in scenarios where the wireless backhaul is considered to be the bottleneck of end-to-end connection. The proposed solution introduces DASHassisting network elements(DANE), assisting signaling, and the assisting algorithm for network-assisted DASH streaming. In experiments with a deployed prototype implementation of the solution in Content Delivery Network(CDN) the number of quality switches reduced by 89% compared to without cache and by 86% compared to without DANE cases. The number of stalls during the playback reduced by 91% compared to without cache and by 93% compared to without DANE. The results show that more constant quality is achieved by the end-users while at the same time the usage of backhaul connection is minimized by utilizing locally available content when applicable.