Scalable video streaming with helper nodes using random linear network coding

Pouya Ostovari, Jie Wu, Abdallah Khreishah, Ness B. Shroff

Research output: Contribution to journalArticlepeer-review

29 Scopus citations

Abstract

Video streaming generates a substantial fraction of the traffic on the Internet. The demands of video streaming also increase the workload on the video server, which in turn leads to substantial slowdowns. In order to resolve the slowdown problem, and to provide a scalable and robust infrastructure to support on-demand streaming, helper-assisted video-on-demand (VoD) systems have been introduced. In this architecture, helper nodes, which are micro-servers with limited storage and bandwidth resources, download and store the user-requested videos from a central server to decrease the load on the central server. Multi-layer videos, in which a video is divided into different layers, can also be used to improve the scalability of the system. In this paper, we study the problem of utilizing the helper nodes to minimize the pressure on the central servers. We formulate the problem as a linear programming using joint inter- And intra-layer network coding. Our solution can also be implemented in a distributed manner. We show how our method can be extended to the case of wireless live streaming, in which a set of videos is broadcasted. Moreover, we extend the proposed method to the case of unreliable connections. We carefully study the convergence and the gain of our distributed approach.

Original languageEnglish (US)
Article number7124543
Pages (from-to)1574-1587
Number of pages14
JournalIEEE/ACM Transactions on Networking
Volume24
Issue number3
DOIs
StatePublished - Jun 1 2016

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Scalable video streaming with helper nodes using random linear network coding'. Together they form a unique fingerprint.

Cite this