计算机科学与探索 ›› 2017, Vol. 11 ›› Issue (7): 1033-1043.DOI: 10.3778/j.issn.1673-9418.1703041

• 学术研究 • 上一篇    下一篇

NDNlive:命名数据网络下的视频直播系统

王丽婧1,2,MOISEENKO Ilya3,何文博4,汪东升2+   

  1. 1. 清华大学 计算机科学与技术系,北京 100084
    2. 清华大学 清华信息科学与技术国家实验室,北京 100084
    3. 加州大学 洛杉矶分校 计算机系,美国 洛杉矶 90095
    4. 清华大学 电子工程系,北京 100084
  • 出版日期:2017-07-01 发布日期:2017-07-07

NDNlive: Live Video Streaming System in Named Data Networking

WANG Lijing1,2, MOISEENKO Ilya3, HE Wenbo4, WANG Dongsheng2+   

  1. 1. Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
    2. Tsinghua National Laboratory for Information Science and Technology, Tsinghua University, Beijing 100084, China
    3. Computer Science Department, University of California, Los Angeles 90095, USA
    4. Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
  • Online:2017-07-01 Published:2017-07-07

摘要: 命名数据网络(named data networking,NDN)通过将IP网络中以地理位置驱动的信息交互方式转变成为以数据为中心的信息交互模式,为内容分发应用例如视频播放提供了更好的支持。通过利用NDN的命名机制与数据获取模式,设计并实现了基于NDN的视频直播系统(NDNlive),将实时捕捉的视频传输给多用户。与传统的定长切片技术不同,NDNlive将视频流按照应用数据单元(帧)进行切分与获取。同时对于音频、视频和元数据信息,依照其数据属性和生成模式采用不同的数据获取方法。由于帧获取流水线策略提供的灵活性,NDNlive可以容忍小的网络问题。NDNlive被部署在NDN全球测试平台中,实验结果表明,NDNlive可以在全球跨11个时区提供流畅和同步的视频直播流。

关键词: 命名数据网络, 视频流, 消费者/生产者编程接口

Abstract: By adopting a data-centric information exchange pattern instead of the IP networks?? location-driven pattern, named data networking (NDN) offers better support for content distribution applications such as video streaming application. This paper proposes NDNlive,?exploiting NDN features such as naming conventions and data retrieval pattern to stream live video to multiple players. Instead of fixed-size segmentation, NDNlive chops video stream into frames which are the application data units (ADU). For the audio, video content and the media metadata, NDNlive uses different data retrieval strategies according to their data properties and different data generation patterns. NDNlive can tolerate small network problems thanks to the flexibility provided by the frame fetch pipelining strategy. It has been deployed over the world-wide NDN Testbed. The experiments show that NDNlive can provide fluent and synchronized video streaming across 11 time zones in the worldwide.

Key words: named data networking, video streaming, Consumer/Producer API