Optimal mode selection for full-duplex enabled D2D cognitive networks

Noman Haider*, Ahsan Ali, Cristo Suarez-Rodriguez, Eryk Dutkiewicz

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

9 Citations (Scopus)


Full-Duplex (FD) and Device-to-Device (D2D) communications have been recognized as one of the successful solutions of spectrum scarcity in 5G networks. Significant advancements in self-interference-to-power-ratio (SIPR) reduction have paved the way for FD use to double the data rates and reduce the latency. This advantage can now be exploited to optimize dynamic spectrum sharing among different radio access technologies in cognitive networks. However, protecting the primary user communication has been a challenging problem in such coexistence. In this paper, we provide an abstract level analysis of protecting primary users reception based on secondary users FD enabled communication. We also propose optimal mode selection (Half-duplex, Full-duplex, or silent) for secondary D2D users depending on its impact on primary users. Our analysis presents the significant advantage of D2D mode selection in terms of efficient spectrum utilization while protecting the primary user transmission, thus, leading the way for FD enabled D2D setup. Depending on the location and transmit power of D2D users, the induced aggregate interference should not violate the interference threshold of primary users. For this, we characterize the interference from D2D links and derive the probability for successful D2D users for half-duplex and full-duplex modes. The analyses are further supported by theoretical and extensive simulation results.

Original languageEnglish
Pages (from-to)57298-57311
Number of pages14
JournalIEEE Access
Publication statusPublished - 2019


  • 5G
  • cognitive networks
  • device-to-device
  • full duplex
  • guard zone
  • interference protection
  • stochastic geometry
  • success probability


Dive into the research topics of 'Optimal mode selection for full-duplex enabled D2D cognitive networks'. Together they form a unique fingerprint.

Cite this