Context-adaptive information fFlow allocation and media delivery in online social networks

Jacob Chakareski, Pascal Frossard

Research output: Contribution to journalArticlepeer-review

9 Scopus citations


This paper investigates context-driven flow allocation and media delivery in online social networks. We exploit information on contacts and content preferences found in social networking applications to provide efficient network services and operation at the underlying transport layer. We formulate a linear programming framework that maximizes the information flowcost ratio of the transport network serving the nodes in the social graph. For practical deployments, we also design a distributed version of the optimization framework that provides similar performance to its centralized counterpart, with lower complexity. In addition, we devise a tracker-based system for efficient content discovery in peer-to-peer (P2P) systems based on social network information. Finally, we design a context-aware packet scheduling technique that maximizes the utility of media delivery among the members of the social network. We provide a comprehensive investigation of the performance of our optimization strategies through both simulations and analysis. We demonstrate their significant advantages over several performance factors relative to conventional solutions that do not employ social network information in their operation.

Original languageEnglish (US)
Article number5456186
Pages (from-to)732-745
Number of pages14
JournalIEEE Journal on Selected Topics in Signal Processing
Issue number4
StatePublished - Aug 2010
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering


  • Context-driven networking
  • flow allocation
  • information flow-cost ratio
  • media delivery
  • online social networks
  • packet scheduling
  • peer-to-peer systems


Dive into the research topics of 'Context-adaptive information fFlow allocation and media delivery in online social networks'. Together they form a unique fingerprint.

Cite this