作者: Srihari Nelakuditi , Ewa Kusmierek , Raja R Harinath , Zhi-Li Zhang
DOI:
关键词:
摘要: In recent years, one of the most popular Internet applications is web-based audio and video playback, where stored streamed from server to a client on-demand. Rigid playback deadlines coupled with constraints on resources such as network bandwidth buffer make delivery challenging task [2]. These could be limited in way that it may not possible deliver full-quality video. situation, desirable minimize degradation quality while operating within resource [9]. Layered encoding proposed provide finer control quality: signal split into layers subset these chosen are met [5]. However trivial select better but consistent ensured when conditions constantly varying. our work, we address this layer selection problem layered show how smoother 1 can provided by utilizing for prefetching. We first define smoothness criteria, design metrics measuring it, then develop off-line algorithms maximize case varying known priori. describe an adaptive algorithm providing smoothed doesn’t assume any knowledge about future availability. The results experiments comparing performance schemes presented. conclude paper brief discussion work.