Interference-aware relay selection scheme for two-hop relay networks with multiple source-destination pairs

Sheng Zhou, Jie Xu, Zhisheng Niu

Research output: Contribution to journalArticlepeer-review

27 Scopus citations


An interference-aware relay selection scheme is proposed for two-hop relay networks with multiple source-destination (S-D) pairs, each of which is assisted by a relay selected from the candidates. We address the interference from relays to their nonintended destinations, and the interference level at any destination is jointly affected by relay selections of all S-D pairs. Upper and lower bounds of the two-hop transmission rate are derived, and the joint impact of the interference from multiple relays is decoupled to each S-D pair. Based on these bounds, an interference-aware relay selection metric for each S-D pair is proposed, which reflects both the relay-assisted two-hop rate and the induced interference to other destinations. The decoupled metric enables us to formulate the relay selection as a weighted bipartite matching problem, with the S-D pairs and the candidate relays as the two sides. Both the centralized scheme based on the Hungarian algorithm and the distributed scheme based on matching games are presented. Simulation results show that the distributed scheme performs nearly the same as the centralized scheme, and both schemes substantially outperform the conventional min-max relay selection scheme in terms of network sum rate.

Original languageEnglish (US)
Article number6408240
Pages (from-to)2327-2338
Number of pages12
JournalIEEE Transactions on Vehicular Technology
Issue number5
StatePublished - 2013
Externally publishedYes


  • Bipartite matching
  • collaborative relay
  • interference limited system
  • relay selection

ASJC Scopus subject areas

  • Automotive Engineering
  • Aerospace Engineering
  • Electrical and Electronic Engineering
  • Applied Mathematics


Dive into the research topics of 'Interference-aware relay selection scheme for two-hop relay networks with multiple source-destination pairs'. Together they form a unique fingerprint.

Cite this