Performance Of Video Adaptive Streaming In Cache-Enabled Networks

Abstract

With the development of Internet, the increasing of users who view video online leads to the necessary of having a standard to watch online video. Dynamic Adaptive Streaming over HTTP (DASH) is one of the most popular choices of streaming video in order to reduce network traffic. Videos are duplicated in the cache nodes which stay between the clients and the servers. However, the intermediate cache node may cause bitrate oscillations because of the difference in the bandwidth from the viewer to the cache node and from that to the server. Therefore, we develop an adaptation method in DASH so as to not only prevent bitrate oscillations but also get higher the performance of video than current methods. Moreover, our proposal combines both throughput-based and buffer-based adaptation methods.