Full-Duplex or Half-Duplex: A Bayesian Game for Wireless Networks with Heterogeneous Self-Interference Cancellation Capabilities
AffiliationUniv Arizona, Dept Elect & Comp Engn
full-duplex wireless networks
MetadataShow full item record
PublisherIEEE COMPUTER SOC
CitationAfifi, W., Abdel-Rahman, M. J., Krunz, M., & MacKenzie, A. B. (2017). Full-duplex or Half-duplex: A Bayesian Game for Wireless Networks with Heterogeneous Self-interference Cancellation Capabilities. IEEE Transactions on Mobile Computing.
Rights© 2017 IEEE.
Collection InformationThis item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at firstname.lastname@example.org.
AbstractRecently, tremendous progress has been made in self-interference cancellation (SIC) techniques that enable a wireless device to transmit and receive data simultaneously on the same frequency channel, a.k.a. in-band full-duplex (FD). Although operating in FD mode significantly improves the throughput of a single wireless link, it doubles the number of concurrent transmissions, which limits the potential for coexistence between multiple FD-enabled links. In this paper, we consider the coexistence problem of concurrent transmissions between multiple FD-enabled links with different SIC capabilities; each link can operate in either FD or half-duplex mode. First, we consider two links and formulate the interactions between them as a Bayesian game. In this game, each link tries to maximize its throughput while minimizing the transmission power cost. We derive a closed-form expression for the Bayesian Nash equilibrium and determine the conditions under which no outage occurs at either link. Then, we study the coexistence problem between more than two links, assuming that each link is only affected by its dominant interfering link. We show that under this assumption, no more than two links will be involved in a single game. Finally, we corroborate our analytical findings via extensive simulations and numerical results.
VersionFinal accepted manuscript
SponsorsUS National Science Foundation [IIP-1265960, CNS-1563655, IIP-1535573]