A Device-to-Device communication-based disaster response network

Javad Zeraatkar Moghaddam, Muhammad Usman, Fabrizio Granelli

Research output: Contribution to journalArticlepeer-review


Device-to-Device (D2D) communication and cognitive radio are among the key technologies that can help to solve the bandwidth bottleneck problem of cellular networks. Such technologies help to maintain the necessary communication between user devices in case of natural disasters and traffic hotspot situations. In this paper, we efficiently exploit the relay selection and cooperative beamforming (CBF) strategies for D2D communication and show that user devices can securely communicate with each other without infrastructure support. We propose a cluster-based hierarchical topology for a multi-hop cooperative relay network, to cover the potentially long distance between the source and the destination. The cluster heads in the proposal act as primary users (PUs) and the cluster members are regarded as secondary users (SUs) or relay nodes. A system model is presented to analyze the problems of relay selection and power allocation in SUs located in different clusters and hierarchical levels. To investigate the efficiency of our proposed model, we define the utilities of PUs employing relay selection and CBF, considering received signal power and latency. We optimize relay selection and beamforming vector by solving a mixed integer non-linear problem. Numerical results demonstrate the efficiency and robustness of our approach.
Original languageEnglish
Pages (from-to)288-298
Number of pages11
JournalIEEE Transactions on Cognitive Communications and Networking
Issue number2
Early online date2 Feb 2018
Publication statusPublished - Jun 2018


  • cognitive radio
  • D2D
  • 5G networks
  • disaster response
  • public safety
  • power gain
  • spectrum efficiency
  • secure communication
  • cooperative beamforming
  • cluster


Dive into the research topics of 'A Device-to-Device communication-based disaster response network'. Together they form a unique fingerprint.

Cite this